Coal is heterogeneous in nature,and thus the characterization of coal is essential before its use for a specific purpose.Thus,the current study aims to develop a machine vision system for automated coal characterizati...Coal is heterogeneous in nature,and thus the characterization of coal is essential before its use for a specific purpose.Thus,the current study aims to develop a machine vision system for automated coal characterizations.The model was calibrated using 80 image samples that are captured for different coal samples in different angles.All the images were captured in RGB color space and converted into five other color spaces(HSI,CMYK,Lab,xyz,Gray)for feature extraction.The intensity component image of HSI color space was further transformed into four frequency components(discrete cosine transform,discrete wavelet transform,discrete Fourier transform,and Gabor filter)for the texture features extraction.A total of 280 image features was extracted and optimized using a step-wise linear regression-based algorithm for model development.The datasets of the optimized features were used as an input for the model,and their respective coal characteristics(analyzed in the laboratory)were used as outputs of the model.The R-squared values were found to be 0.89,0.92,0.92,and 0.84,respectively,for fixed carbon,ash content,volatile matter,and moisture content.The performance of the proposed artificial neural network model was also compared with the performances of performances of Gaussian process regression,support vector regression,and radial basis neural network models.The study demonstrates the potential of the machine vision system in automated coal characterization.展开更多
This study assessed the feasibility of developing a machine vision system equipped with ultraviolet (UV) light, using changes in fish-surface color to predict aerobic plate count (APC, a standard freshness indicator) ...This study assessed the feasibility of developing a machine vision system equipped with ultraviolet (UV) light, using changes in fish-surface color to predict aerobic plate count (APC, a standard freshness indicator) during storage. The APC values were tested and images of the fish surface were taken when fish were stored at room temperature. Then, images</span></span><span style="font-family:Verdana;"><span style="font-family:Verdana;"><span style="font-family:Verdana;">’</span></span></span><span><span><span><span> color-space conversion among RGB, HSV, and L*a*b* color spaces was carried out and analyzed. The results revealed that a* and b* values from the UV-light image decreased linearly during storage. A further regression analysis of these two parameters with APC value demonstrated a good exponential relationship between the a* value and the APC value (R</span><sup><span>2</span></sup><span> = 0.97), followed by the b* (R</span><sup><span>2</span></sup><span> = 0.85). Therefore, our results suggest that the change in color of the fish surface under UV light can be used to assess fish freshness during storage.展开更多
With the development of wood industry, the processing of wood products becomemore significant. This paper discusses the developmen of machine vision system used to inspect andclassny the various types of defects of wo...With the development of wood industry, the processing of wood products becomemore significant. This paper discusses the developmen of machine vision system used to inspect andclassny the various types of defects of wood suxface. The surface defeds means the variations ofcolour and textUre. The machine vision system is to dated undesirable 'defecs' that can appear onthe surface of rough wood lwnber. A neural network was used within the Blackboard framework fora labeling verification step of the high-level recognition module of vision system. The system hasbere successfully tested on a number of boards from several different species.展开更多
Prompt radiation emitted during accelerator operation poses a significant health risk,necessitating a thorough search and securing of hazardous areas prior to initiation.Currently,manual sweep methods are employed.How...Prompt radiation emitted during accelerator operation poses a significant health risk,necessitating a thorough search and securing of hazardous areas prior to initiation.Currently,manual sweep methods are employed.However,the limitations of manual sweeps have become increasingly evident with the implementation of large-scale accelerators.By leveraging advancements in machine vision technology,the automatic identification of stranded personnel in controlled areas through camera imagery presents a viable solution for efficient search and security.Given the criticality of personal safety for stranded individuals,search and security processes must be sufficiently reliable.To ensure comprehensive coverage,180°camera groups were strategically positioned on both sides of the accelerator tunnel to eliminate blind spots within the monitoring range.The YOLOV8 network model was modified to enable the detection of small targets,such as hands and feet,as well as larger targets formed by individuals near the cameras.Furthermore,the system incorporates a pedestrian recognition model that detects human body parts,and an information fusion strategy is used to integrate the detected head,hands,and feet with the identified pedestrians as a cohesive unit.This strategy enhanced the capability of the model to identify pedestrians obstructed by equipment,resulting in a notable improvement in the recall rate.Specifically,recall rates of 0.915 and 0.82were obtained for Datasets 1 and 2,respectively.Although there was a slight decrease in accuracy,it aligned with the intended purpose of the search-and-secure software design.Experimental tests conducted within an accelerator tunnel demonstrated the effectiveness of this approach in achieving reliable recognition outcomes.展开更多
Machine vision systems(MVSs)are an important component of intelligent systems,such as autonomous vehicles and robots.However,with the continuous increase in data and new application scenarios,new requirements are put ...Machine vision systems(MVSs)are an important component of intelligent systems,such as autonomous vehicles and robots.However,with the continuous increase in data and new application scenarios,new requirements are put forward for the next generation of MVS.There is an urgent need to find new material systems to complement the existing semiconductor technology based on thin-film materials,and new architectures must be explored to improve efficiency.Because of their unique physical properties,two-dimensional(2D)materials have received extensive attention for use in MVSs,especially in biomimetic ones:the human visual system,which can process complex visual information with low power consumption,provides a model for next-generation MVSs.This review paper summarizes the progress and challenges of applying 2D material photodetectors in sense-memory-computational integration and biomimetic image sensors for machine vision.展开更多
Monitring pest populations in paddy fields is important to effectively implement integrated pest management.Light traps are widely used to monitor field pests all over the world.Most conventional light traps still inv...Monitring pest populations in paddy fields is important to effectively implement integrated pest management.Light traps are widely used to monitor field pests all over the world.Most conventional light traps still involve manual identification of target pests from lots of trapped insects,which is time-consuming,labor-intensive and error-prone,especially in pest peak periods.In this paper,we developed an automatic monitoring system for rice light-trap pests based on machine vision.This system is composed of an itelligent light trap,a computer or mobile phone client platform and a cloud server.The light trap firstly traps,kills and disperses insects,then collects images of trapped insects and sends each image to the cloud server.Five target pests in images are automatically identifed and counted by pest identification models loaded in the server.To avoid light-trap insects piling up,a vibration plate and a moving rotation conveyor belt are adopted to disperse these trapped insects.There was a close correlation(r=0.92)between our automatic and manual identification methods based on the daily pest number of one-year images from one light trap.Field experiments demonstrated the effectiveness and accuracy of our automatic light trap monitoring system.展开更多
Food processing companies pursue the distribution of ingredientsthat were packaged according to a certain weight. Particularly, foods like fishare highly demanded and supplied. However, despite the high quantity offis...Food processing companies pursue the distribution of ingredientsthat were packaged according to a certain weight. Particularly, foods like fishare highly demanded and supplied. However, despite the high quantity offish to be supplied, most seafood processing companies have yet to installautomation equipment. Such absence of automation equipment for seafoodprocessing incurs a considerable cost regarding labor force, economy, andtime. Moreover, workers responsible for fish processing are exposed to risksbecause fish processing tasks require the use of dangerous tools, such aspower saws or knives. To solve these problems observed in the fish processingfield, this study proposed a fish cutting point prediction method based onAI machine vision and target weight. The proposed method performs threedimensional(3D) modeling of a fish’s form based on image processing techniquesand partitioned random sample consensus (RANSAC) and extracts 3Dfeature information. Then, it generates a neural network model for predictingfish cutting points according to the target weight by performing machinelearning of the extracted 3D feature information and measured weight information.This study allows for the direct cutting of fish based on cutting pointspredicted by the proposed method. Subsequently, we compared the measuredweight of the cut pieces with the target weight. The comparison result verifiedthat the proposed method showed a mean error rate of approximately 3%.展开更多
Estimation of fruit size in tree fruit crops is essential for selective robotic harvesting and crop-load estimation.Machine vision systems for fruit detection and localization have been studied widely for robotic harv...Estimation of fruit size in tree fruit crops is essential for selective robotic harvesting and crop-load estimation.Machine vision systems for fruit detection and localization have been studied widely for robotic harvesting and crop-load estimation.However,only a few studies have been carried out to estimate fruit size in orchards using machine vision systems.This study was carried out to develop a machine vision system consisting of a color CCD camera and a time-of-flight(TOF)light-based 3D camera for estimating apple size in tree canopies.As a measure of fruit size,the major axis(longest axis)was estimated based on(i)the 3D coordinates of pixels on corresponding apple surfaces,and(ii)the 2D size of individual pixels within apple surfaces.In the 3D coordinates-based method,the distance between pairs of pixels within apple regions were calculated using 3D coordinates,and the maximum distance between all pixel pairs within an apple region was estimated to be the major axis.The accuracy of estimating the major axis using 3D coordinates was 69.1%.In the pixel-size-based method,the physical sizes of pixels were estimated using a calibration model developed based on pixel coordinates and the distance to pixels from the camera.The major axis length was then estimated by summing the size of individual pixels along the major axis of the fruit.The accuracy of size estimation increased to 84.8%when the pixel size-based method was used.The results showed the potential for estimating fruit size in outdoor environments using a 3D machine vision system.展开更多
The inter-class face classification problem is more reasonable than the intra-class classification problem.To address this issue,we have carried out empirical research on classifying Indian people to their geographica...The inter-class face classification problem is more reasonable than the intra-class classification problem.To address this issue,we have carried out empirical research on classifying Indian people to their geographical regions.This work aimed to construct a computational classification model for classifying Indian regional face images acquired from south and east regions of India,referring to human vision.We have created an Automated Human Intelligence System(AHIS)to evaluate human visual capabilities.Analysis of AHIS response showed that face shape is a discriminative feature among the other facial features.We have developed a modified convolutional neural network to characterize the human vision response to improve face classification accuracy.The proposed model achieved mean F1 and Matthew Correlation Coefficient(MCC)of 0.92 and 0.84,respectively,on the validation set,outperforming the traditional Convolutional Neural Network(CNN).The CNN-Contoured Face(CNN-FC)model is developed to train contoured face images to investigate the influence of face shape.Finally,to cross-validate the accuracy of these models,the traditional CNN model is trained on the same dataset.With an accuracy of 92.98%,the Modified-CNN(M-CNN)model has demonstrated that the proposed method could facilitate the tangible impact in intra-classification problems.A novel Indian regional face dataset is created for supporting this supervised classification work,and it will be available to the research community.展开更多
The quality of the stator winding coil directly affects the performance of the motor.A dual-camera online machine vision detection method to detect whether the coil leads and winding regions were qualified was designe...The quality of the stator winding coil directly affects the performance of the motor.A dual-camera online machine vision detection method to detect whether the coil leads and winding regions were qualified was designed.A vision detection platform was designed to capture individual winding images,and an image processing algorithm was used for image pre-processing,template matching and positioning of the coil lead area to set up a coordinate system.After eliminating image noise by Blob analysis,the improved Canny algorithm was used to detect the location of the coil lead paint stripped region,and the time was reduced by about half compared to the Canny algorithm.The coil winding region was trained with the ShuffleNet V2-YOLOv5s model for the dataset,and the detect file was converted to the Open Neural Network Exchange(ONNX)model for the detection of winding cross features with an average accuracy of 99.0%.The software interface of the detection system was designed to perform qualified discrimination tests on the workpieces,and the detection data were recorded and statistically analyzed.The results showed that the stator winding coil qualified discrimination accuracy reached 96.2%,and the average detection time of a single workpiece was about 300 ms,while YOLOv5s took less than 30 ms.展开更多
Computer vision(CV)was developed for computers and other systems to act or make recommendations based on visual inputs,such as digital photos,movies,and other media.Deep learning(DL)methods are more successful than ot...Computer vision(CV)was developed for computers and other systems to act or make recommendations based on visual inputs,such as digital photos,movies,and other media.Deep learning(DL)methods are more successful than other traditional machine learning(ML)methods inCV.DL techniques can produce state-of-the-art results for difficult CV problems like picture categorization,object detection,and face recognition.In this review,a structured discussion on the history,methods,and applications of DL methods to CV problems is presented.The sector-wise presentation of applications in this papermay be particularly useful for researchers in niche fields who have limited or introductory knowledge of DL methods and CV.This review will provide readers with context and examples of how these techniques can be applied to specific areas.A curated list of popular datasets and a brief description of them are also included for the benefit of readers.展开更多
Text detection and recognition is a hot topic in computer vision,which is considered to be the further development of the traditional optical character recognition(OCR)technology.With the rapid development of machine ...Text detection and recognition is a hot topic in computer vision,which is considered to be the further development of the traditional optical character recognition(OCR)technology.With the rapid development of machine vision system and the wide application of deep learning algorithms,text recognition has achieved excellent performance.In contrast,detecting text block from complex natural scenes is still a challenging task.At present,many advanced natural scene text detection algorithms have been proposed,but most of them run slow due to the complexity of the detection pipeline and can not be applied to industrial scenes.In this paper,we proposed a CCD based machine vision system for realtime text detection in invoice images.In this system,we applied optimizations from several aspects including the optical system,the hardware architecture,and the deep learning algorithm to improve the speed performance of the machine vision system.The experimental data confirms that the optimization methods can significantly improve the running speed of the machine vision system and make it meeting the real-time text detection requirements in industrial scenarios.展开更多
In this paper, an automatic inspection system for weld surface appearance using machine vision has been developed to recognize weld surface defects such as porosities, cracks, etc. It can replace conventional manual v...In this paper, an automatic inspection system for weld surface appearance using machine vision has been developed to recognize weld surface defects such as porosities, cracks, etc. It can replace conventional manual visual inspection method, which is tedious, time-consuming, subjective, experience-depended, and sometimes biased. The system consists of a CCD camera, a self-designed annular light source, a sensor controller, a frame grabbing card, a computer and so on. After acquiring weld surface appearance images using CCD, the images are preprocessed using median filtering and a series of image enhancement algorithms. Then a dynamic threshold and morphology algorithms are applied to segment defect object. Finally, defect features information is obtained by eight neighborhoods boundary chain code algorithm. Experimental results show that the developed system is capable of inspecting most surface defects such as porosities, cracks with high reliability and accuracy.展开更多
To make the problems of existing high requirements of calibration tools, complex global calibration process addressed for monocular multi-view visual measurement system during measurement, in the paper, a global calib...To make the problems of existing high requirements of calibration tools, complex global calibration process addressed for monocular multi-view visual measurement system during measurement, in the paper, a global calibration method is proposed for the geometric properties of rotational correlation motion and the absolute orientation of the field of view without over lap. Firstly, a dual-camera system is constructed for photographing and collecting the rotating image sequence of two flat targets rigidly connected by a long rod at different positions, and based on the known parameters, such as, target feature image, world coordinates, camera internal parameters and so on, then the global PnP optimization method is used to solve the rotation axis and the reference point at different positions;Then, the absolute orientation matrix is constructed based on the parameters of rotation axis, reference point and connecting rod length obtained by this method. In the end, the singular value decomposition method is used to find the optimal rotation matrix, and then get the translation matrix. It’s shown based on simulation and actual tests that in comparison with the existing methods, the maximum attitude and pose errors is 0.0083˚ and 0.3657 mm, respectively, which improves the accuracy by 27.8% and 24.4%, respectively. The calibration device in this paper is simple, and there are no parallel, vertical and coplanar requirements between multiple rotating positions. At the same time, in view of the calibration accuracy, the accuracy requirements of most application scenarios can be met.展开更多
Tubes are used widely in aerospace vehicles, and their accurate assembly can directly affect the assembling reliability and the quality of products. It is important to measure the processed tube's endpoints and then ...Tubes are used widely in aerospace vehicles, and their accurate assembly can directly affect the assembling reliability and the quality of products. It is important to measure the processed tube's endpoints and then fix any geometric errors correspondingly. However, the traditional tube inspection method is time-consuming and complex operations. Therefore, a new measurement method for a tube's endpoints based on machine vision is proposed. First, reflected light on tube's surface can be removed by using photometric linearization. Then, based on the optimization model for the tube's endpoint measurements and the principle of stereo matching, the global coordinates and the relative distance of the tube's endpoint are obtained. To confirm the feasibility, ll tubes are processed to remove the reflected light and then the endpoint's positions of tubes are measured. The experiment results show that the measurement repeatability accuracy is 0.167 mm, and the absolute accuracy is 0.328 ram. The measurement takes less than 1 min. The proposed method based on machine vision can measure the tube's endpoints without any surface treatment or any tools and can realize on line measurement.展开更多
This research aimed to improve selection of pepper seeds for separating high-quality seeds from low-quality seeds. Past research has shown that seed vigor is significantly related to the seed color and size, thus seve...This research aimed to improve selection of pepper seeds for separating high-quality seeds from low-quality seeds. Past research has shown that seed vigor is significantly related to the seed color and size, thus several physical features were identified as candidate predictors of high seed quality. Image recognition software was used to automate recognition of seed feature quality using 400 kernels of pepper cultivar 101. In addition, binary logistic regression and a neural network were applied to determine models with high predictive value of seed germination. Single-kernel germination tests were conducted to validate the predictive value of the identified features. The best predictors of seed vigor were determined by the highest correlation observed between the physical features and the subsequent fresh weight of seedlings that germinated from the 400 seeds. Correlation analysis showed that fresh weight was significantly positively correlated with eight physical features: three color features (R, a*, brightness), width, length, projected area, and single-kernel density, and weight. In contrast, fresh weight significantly negatively correlated with the feature of hue. In analyses of two of the highest correlating single features,' germination percentage increased from 59.3 to 71.8% when a*〉3, and selection rate peaked at 57.8%. Germination percentage increased from 59.3 to 79.4%, and the selection rate reached 76.8%, when single-kernel weight 〉0.0064 g. The most effective model was based on a multilayer perceptron (MLP) neural network, consisting of 15 physical traits as variables, and a stability calculated as 99.4%. Germination percentage in a calibration set of seeds was 79.1% and the selection rate was 90.0%. These results indicated that the model was effective in predicting seed germination based on physical features and could be used as a guide for quality control in seed selection. Automated systems based on machine vision and model classifiers can contribute to reducing the costs and labor required in the selection of pepper seeds.展开更多
When characterizing flows in miniaturized channels, the determination of the dynamic contact angle is important. By measuring the dynamic contact angle, the flow properties of the flowing liquid and the effect of mate...When characterizing flows in miniaturized channels, the determination of the dynamic contact angle is important. By measuring the dynamic contact angle, the flow properties of the flowing liquid and the effect of material properties on the flow can be characterized. A machine vision based system to measure the contact angle of front or rear menisci of a moving liquid plug is described in this article. In this research, transparent flow channels fabricated on thermoplastic polymer and sealed with an adhesive tape are used. The transparency of the channels enables image based monitoring and measurement of flow variables, including the dynamic contact angle. It is shown that the dynamic angle can be measured from a liquid flow in a channel using the image based measurement system. An image processing algorithm has been developed in a MATLAB environment. Images are taken using a CCD camera and the channels are illuminated using a custom made ring light. Two fitting methods, a circle and two parabolas, are experimented and the results are compared in the measurement of the dynamic contact angles.展开更多
In this study, we established a dynamic morphological model using the accumulated thermal effectiveness and photosynthetic active radiation (PAR) (A-TEP), aiming to explore the relationship between muskmelon (Cuc...In this study, we established a dynamic morphological model using the accumulated thermal effectiveness and photosynthetic active radiation (PAR) (A-TEP), aiming to explore the relationship between muskmelon (Cucumis melo L.) fruit attributes and environmental factors. Muskmelon surface color was described by parameters of red, green, blue, hue, saturation and brightness (HSI). Three characteristic parameters, gray level co-occurrence matrix (GLCM), angular second moment (ASM), entropy, contrast, and the coverage rate were used to describe the process of muskmelon fruit netting formation. ASM was not significant difference during muskmelon fruit growth. The number and deep of netting stripes gradually increased with fruit growth. Coverage rate increased rapidly for 15-30 d after pollination. The vertical and horizontal diameters of muskmelon fruit were followed a logistic curve. And root mean squared errors (RMSE) between the simulated and measured vertical and horizontal diameters were 3.527 and 4.696 mm, respectively. RMSE of red, green, blue, saturation and brightness were 0.999, 2.690, 2.992, 0.033 and 5.51, respectively, and the RMSE for entropy, contrast and coverage rates were 0.077, 0.063 and 0.015, respectively, indicating a well consistent between measured and simulated values.展开更多
The objective of this study was to develop an online tool-wear-measurement scheme for small diameter end-mills based on machine vision to increase tool life and the production efficiency. The geometrical features of w...The objective of this study was to develop an online tool-wear-measurement scheme for small diameter end-mills based on machine vision to increase tool life and the production efficiency. The geometrical features of wear zone of each end mill were analyzed, and three tool wear criterions of small-diameter end mills were defined. With the uEye camera, macro lens and 3-axis micro milling machine, it was proved the feasibility of measuring flank wear with the milling tests on a 45# steel workpiece. The design of experiment (DOE) showed that Vc was the most remarkable effect factor for the flank wear of small-diameter end mill. The wear curve of the experiments of milling was very similar to the Taylor curve.展开更多
Manual monitoring and seam tracking through watching weld pool images in real-time, by naked eyes or by industrial TV, are experience-depended, subjective, labor intensive, and sometimes biased. So it is necessary to ...Manual monitoring and seam tracking through watching weld pool images in real-time, by naked eyes or by industrial TV, are experience-depended, subjective, labor intensive, and sometimes biased. So it is necessary to realize the automation of computer-aided seam tracking. A PAW (plasma arc welding) seam tracking system was developed, which senses the molten pool and the seam in one frame by a vision sensor, and then detects the seam deviation to adjust the work piece motion adaptively to the seam position sensed by vision sensor. A novel molten pool area image-processing algorithm based on machine vision was proposed. The algorithm processes each image at the speed of 20 frames/second in real-time to extract three feature variables to get the seam deviation. It is proved experimentally that the algorithm is very fast and effective. Issues related to the algorithm are also discussed.展开更多
文摘Coal is heterogeneous in nature,and thus the characterization of coal is essential before its use for a specific purpose.Thus,the current study aims to develop a machine vision system for automated coal characterizations.The model was calibrated using 80 image samples that are captured for different coal samples in different angles.All the images were captured in RGB color space and converted into five other color spaces(HSI,CMYK,Lab,xyz,Gray)for feature extraction.The intensity component image of HSI color space was further transformed into four frequency components(discrete cosine transform,discrete wavelet transform,discrete Fourier transform,and Gabor filter)for the texture features extraction.A total of 280 image features was extracted and optimized using a step-wise linear regression-based algorithm for model development.The datasets of the optimized features were used as an input for the model,and their respective coal characteristics(analyzed in the laboratory)were used as outputs of the model.The R-squared values were found to be 0.89,0.92,0.92,and 0.84,respectively,for fixed carbon,ash content,volatile matter,and moisture content.The performance of the proposed artificial neural network model was also compared with the performances of performances of Gaussian process regression,support vector regression,and radial basis neural network models.The study demonstrates the potential of the machine vision system in automated coal characterization.
文摘This study assessed the feasibility of developing a machine vision system equipped with ultraviolet (UV) light, using changes in fish-surface color to predict aerobic plate count (APC, a standard freshness indicator) during storage. The APC values were tested and images of the fish surface were taken when fish were stored at room temperature. Then, images</span></span><span style="font-family:Verdana;"><span style="font-family:Verdana;"><span style="font-family:Verdana;">’</span></span></span><span><span><span><span> color-space conversion among RGB, HSV, and L*a*b* color spaces was carried out and analyzed. The results revealed that a* and b* values from the UV-light image decreased linearly during storage. A further regression analysis of these two parameters with APC value demonstrated a good exponential relationship between the a* value and the APC value (R</span><sup><span>2</span></sup><span> = 0.97), followed by the b* (R</span><sup><span>2</span></sup><span> = 0.85). Therefore, our results suggest that the change in color of the fish surface under UV light can be used to assess fish freshness during storage.
文摘With the development of wood industry, the processing of wood products becomemore significant. This paper discusses the developmen of machine vision system used to inspect andclassny the various types of defects of wood suxface. The surface defeds means the variations ofcolour and textUre. The machine vision system is to dated undesirable 'defecs' that can appear onthe surface of rough wood lwnber. A neural network was used within the Blackboard framework fora labeling verification step of the high-level recognition module of vision system. The system hasbere successfully tested on a number of boards from several different species.
文摘Prompt radiation emitted during accelerator operation poses a significant health risk,necessitating a thorough search and securing of hazardous areas prior to initiation.Currently,manual sweep methods are employed.However,the limitations of manual sweeps have become increasingly evident with the implementation of large-scale accelerators.By leveraging advancements in machine vision technology,the automatic identification of stranded personnel in controlled areas through camera imagery presents a viable solution for efficient search and security.Given the criticality of personal safety for stranded individuals,search and security processes must be sufficiently reliable.To ensure comprehensive coverage,180°camera groups were strategically positioned on both sides of the accelerator tunnel to eliminate blind spots within the monitoring range.The YOLOV8 network model was modified to enable the detection of small targets,such as hands and feet,as well as larger targets formed by individuals near the cameras.Furthermore,the system incorporates a pedestrian recognition model that detects human body parts,and an information fusion strategy is used to integrate the detected head,hands,and feet with the identified pedestrians as a cohesive unit.This strategy enhanced the capability of the model to identify pedestrians obstructed by equipment,resulting in a notable improvement in the recall rate.Specifically,recall rates of 0.915 and 0.82were obtained for Datasets 1 and 2,respectively.Although there was a slight decrease in accuracy,it aligned with the intended purpose of the search-and-secure software design.Experimental tests conducted within an accelerator tunnel demonstrated the effectiveness of this approach in achieving reliable recognition outcomes.
基金supported by the National Natural Science Foundation of China(Grant Nos.61905266,62004207,61904184,62005303,62175045,62134009)Special grants from China Post-doctoral Science Foundation(Grant No.2021M700156)+1 种基金Youth Innovation Promotion Association CAS,Hangzhou Key Research and Development Program(Grant No.20212013B01)the Science and Technology Commission of Shanghai Municipality(Grant No.21JC1406100 and 20YF1455900).
文摘Machine vision systems(MVSs)are an important component of intelligent systems,such as autonomous vehicles and robots.However,with the continuous increase in data and new application scenarios,new requirements are put forward for the next generation of MVS.There is an urgent need to find new material systems to complement the existing semiconductor technology based on thin-film materials,and new architectures must be explored to improve efficiency.Because of their unique physical properties,two-dimensional(2D)materials have received extensive attention for use in MVSs,especially in biomimetic ones:the human visual system,which can process complex visual information with low power consumption,provides a model for next-generation MVSs.This review paper summarizes the progress and challenges of applying 2D material photodetectors in sense-memory-computational integration and biomimetic image sensors for machine vision.
基金Supported by the Fundamental Public Welfare Research Program of Zhejiang Provincial Natural Science Foundation,China(LGN18C140007 and Y20C140024)the National High Technology Research and Development Program of China(863 Program,2013AA102402)the Agricultural Science and Technology Innovation Program of Chinese Academy of Agricultural Sciences.
文摘Monitring pest populations in paddy fields is important to effectively implement integrated pest management.Light traps are widely used to monitor field pests all over the world.Most conventional light traps still involve manual identification of target pests from lots of trapped insects,which is time-consuming,labor-intensive and error-prone,especially in pest peak periods.In this paper,we developed an automatic monitoring system for rice light-trap pests based on machine vision.This system is composed of an itelligent light trap,a computer or mobile phone client platform and a cloud server.The light trap firstly traps,kills and disperses insects,then collects images of trapped insects and sends each image to the cloud server.Five target pests in images are automatically identifed and counted by pest identification models loaded in the server.To avoid light-trap insects piling up,a vibration plate and a moving rotation conveyor belt are adopted to disperse these trapped insects.There was a close correlation(r=0.92)between our automatic and manual identification methods based on the daily pest number of one-year images from one light trap.Field experiments demonstrated the effectiveness and accuracy of our automatic light trap monitoring system.
基金supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF)funded by the Ministry of Education (NRF-2020R1I1A3073313).
文摘Food processing companies pursue the distribution of ingredientsthat were packaged according to a certain weight. Particularly, foods like fishare highly demanded and supplied. However, despite the high quantity offish to be supplied, most seafood processing companies have yet to installautomation equipment. Such absence of automation equipment for seafoodprocessing incurs a considerable cost regarding labor force, economy, andtime. Moreover, workers responsible for fish processing are exposed to risksbecause fish processing tasks require the use of dangerous tools, such aspower saws or knives. To solve these problems observed in the fish processingfield, this study proposed a fish cutting point prediction method based onAI machine vision and target weight. The proposed method performs threedimensional(3D) modeling of a fish’s form based on image processing techniquesand partitioned random sample consensus (RANSAC) and extracts 3Dfeature information. Then, it generates a neural network model for predictingfish cutting points according to the target weight by performing machinelearning of the extracted 3D feature information and measured weight information.This study allows for the direct cutting of fish based on cutting pointspredicted by the proposed method. Subsequently, we compared the measuredweight of the cut pieces with the target weight. The comparison result verifiedthat the proposed method showed a mean error rate of approximately 3%.
基金supported in part by the USDA’s Hatch and Multistate Project Funds(Accession Nos.1005756 and 1001246)。
文摘Estimation of fruit size in tree fruit crops is essential for selective robotic harvesting and crop-load estimation.Machine vision systems for fruit detection and localization have been studied widely for robotic harvesting and crop-load estimation.However,only a few studies have been carried out to estimate fruit size in orchards using machine vision systems.This study was carried out to develop a machine vision system consisting of a color CCD camera and a time-of-flight(TOF)light-based 3D camera for estimating apple size in tree canopies.As a measure of fruit size,the major axis(longest axis)was estimated based on(i)the 3D coordinates of pixels on corresponding apple surfaces,and(ii)the 2D size of individual pixels within apple surfaces.In the 3D coordinates-based method,the distance between pairs of pixels within apple regions were calculated using 3D coordinates,and the maximum distance between all pixel pairs within an apple region was estimated to be the major axis.The accuracy of estimating the major axis using 3D coordinates was 69.1%.In the pixel-size-based method,the physical sizes of pixels were estimated using a calibration model developed based on pixel coordinates and the distance to pixels from the camera.The major axis length was then estimated by summing the size of individual pixels along the major axis of the fruit.The accuracy of size estimation increased to 84.8%when the pixel size-based method was used.The results showed the potential for estimating fruit size in outdoor environments using a 3D machine vision system.
文摘The inter-class face classification problem is more reasonable than the intra-class classification problem.To address this issue,we have carried out empirical research on classifying Indian people to their geographical regions.This work aimed to construct a computational classification model for classifying Indian regional face images acquired from south and east regions of India,referring to human vision.We have created an Automated Human Intelligence System(AHIS)to evaluate human visual capabilities.Analysis of AHIS response showed that face shape is a discriminative feature among the other facial features.We have developed a modified convolutional neural network to characterize the human vision response to improve face classification accuracy.The proposed model achieved mean F1 and Matthew Correlation Coefficient(MCC)of 0.92 and 0.84,respectively,on the validation set,outperforming the traditional Convolutional Neural Network(CNN).The CNN-Contoured Face(CNN-FC)model is developed to train contoured face images to investigate the influence of face shape.Finally,to cross-validate the accuracy of these models,the traditional CNN model is trained on the same dataset.With an accuracy of 92.98%,the Modified-CNN(M-CNN)model has demonstrated that the proposed method could facilitate the tangible impact in intra-classification problems.A novel Indian regional face dataset is created for supporting this supervised classification work,and it will be available to the research community.
基金National Natural Science Foundation of China(No.U1831123)。
文摘The quality of the stator winding coil directly affects the performance of the motor.A dual-camera online machine vision detection method to detect whether the coil leads and winding regions were qualified was designed.A vision detection platform was designed to capture individual winding images,and an image processing algorithm was used for image pre-processing,template matching and positioning of the coil lead area to set up a coordinate system.After eliminating image noise by Blob analysis,the improved Canny algorithm was used to detect the location of the coil lead paint stripped region,and the time was reduced by about half compared to the Canny algorithm.The coil winding region was trained with the ShuffleNet V2-YOLOv5s model for the dataset,and the detect file was converted to the Open Neural Network Exchange(ONNX)model for the detection of winding cross features with an average accuracy of 99.0%.The software interface of the detection system was designed to perform qualified discrimination tests on the workpieces,and the detection data were recorded and statistically analyzed.The results showed that the stator winding coil qualified discrimination accuracy reached 96.2%,and the average detection time of a single workpiece was about 300 ms,while YOLOv5s took less than 30 ms.
基金supported by the Project SP2023/074 Application of Machine and Process Control Advanced Methods supported by the Ministry of Education,Youth and Sports,Czech Republic.
文摘Computer vision(CV)was developed for computers and other systems to act or make recommendations based on visual inputs,such as digital photos,movies,and other media.Deep learning(DL)methods are more successful than other traditional machine learning(ML)methods inCV.DL techniques can produce state-of-the-art results for difficult CV problems like picture categorization,object detection,and face recognition.In this review,a structured discussion on the history,methods,and applications of DL methods to CV problems is presented.The sector-wise presentation of applications in this papermay be particularly useful for researchers in niche fields who have limited or introductory knowledge of DL methods and CV.This review will provide readers with context and examples of how these techniques can be applied to specific areas.A curated list of popular datasets and a brief description of them are also included for the benefit of readers.
文摘Text detection and recognition is a hot topic in computer vision,which is considered to be the further development of the traditional optical character recognition(OCR)technology.With the rapid development of machine vision system and the wide application of deep learning algorithms,text recognition has achieved excellent performance.In contrast,detecting text block from complex natural scenes is still a challenging task.At present,many advanced natural scene text detection algorithms have been proposed,but most of them run slow due to the complexity of the detection pipeline and can not be applied to industrial scenes.In this paper,we proposed a CCD based machine vision system for realtime text detection in invoice images.In this system,we applied optimizations from several aspects including the optical system,the hardware architecture,and the deep learning algorithm to improve the speed performance of the machine vision system.The experimental data confirms that the optimization methods can significantly improve the running speed of the machine vision system and make it meeting the real-time text detection requirements in industrial scenarios.
文摘In this paper, an automatic inspection system for weld surface appearance using machine vision has been developed to recognize weld surface defects such as porosities, cracks, etc. It can replace conventional manual visual inspection method, which is tedious, time-consuming, subjective, experience-depended, and sometimes biased. The system consists of a CCD camera, a self-designed annular light source, a sensor controller, a frame grabbing card, a computer and so on. After acquiring weld surface appearance images using CCD, the images are preprocessed using median filtering and a series of image enhancement algorithms. Then a dynamic threshold and morphology algorithms are applied to segment defect object. Finally, defect features information is obtained by eight neighborhoods boundary chain code algorithm. Experimental results show that the developed system is capable of inspecting most surface defects such as porosities, cracks with high reliability and accuracy.
文摘To make the problems of existing high requirements of calibration tools, complex global calibration process addressed for monocular multi-view visual measurement system during measurement, in the paper, a global calibration method is proposed for the geometric properties of rotational correlation motion and the absolute orientation of the field of view without over lap. Firstly, a dual-camera system is constructed for photographing and collecting the rotating image sequence of two flat targets rigidly connected by a long rod at different positions, and based on the known parameters, such as, target feature image, world coordinates, camera internal parameters and so on, then the global PnP optimization method is used to solve the rotation axis and the reference point at different positions;Then, the absolute orientation matrix is constructed based on the parameters of rotation axis, reference point and connecting rod length obtained by this method. In the end, the singular value decomposition method is used to find the optimal rotation matrix, and then get the translation matrix. It’s shown based on simulation and actual tests that in comparison with the existing methods, the maximum attitude and pose errors is 0.0083˚ and 0.3657 mm, respectively, which improves the accuracy by 27.8% and 24.4%, respectively. The calibration device in this paper is simple, and there are no parallel, vertical and coplanar requirements between multiple rotating positions. At the same time, in view of the calibration accuracy, the accuracy requirements of most application scenarios can be met.
基金Supported by National Natural Science Foundation of China(Grant No51305031)
文摘Tubes are used widely in aerospace vehicles, and their accurate assembly can directly affect the assembling reliability and the quality of products. It is important to measure the processed tube's endpoints and then fix any geometric errors correspondingly. However, the traditional tube inspection method is time-consuming and complex operations. Therefore, a new measurement method for a tube's endpoints based on machine vision is proposed. First, reflected light on tube's surface can be removed by using photometric linearization. Then, based on the optimization model for the tube's endpoint measurements and the principle of stereo matching, the global coordinates and the relative distance of the tube's endpoint are obtained. To confirm the feasibility, ll tubes are processed to remove the reflected light and then the endpoint's positions of tubes are measured. The experiment results show that the measurement repeatability accuracy is 0.167 mm, and the absolute accuracy is 0.328 ram. The measurement takes less than 1 min. The proposed method based on machine vision can measure the tube's endpoints without any surface treatment or any tools and can realize on line measurement.
基金supported by the Beijing Municipal Science and Technology Project,China (Z151100001015004)
文摘This research aimed to improve selection of pepper seeds for separating high-quality seeds from low-quality seeds. Past research has shown that seed vigor is significantly related to the seed color and size, thus several physical features were identified as candidate predictors of high seed quality. Image recognition software was used to automate recognition of seed feature quality using 400 kernels of pepper cultivar 101. In addition, binary logistic regression and a neural network were applied to determine models with high predictive value of seed germination. Single-kernel germination tests were conducted to validate the predictive value of the identified features. The best predictors of seed vigor were determined by the highest correlation observed between the physical features and the subsequent fresh weight of seedlings that germinated from the 400 seeds. Correlation analysis showed that fresh weight was significantly positively correlated with eight physical features: three color features (R, a*, brightness), width, length, projected area, and single-kernel density, and weight. In contrast, fresh weight significantly negatively correlated with the feature of hue. In analyses of two of the highest correlating single features,' germination percentage increased from 59.3 to 71.8% when a*〉3, and selection rate peaked at 57.8%. Germination percentage increased from 59.3 to 79.4%, and the selection rate reached 76.8%, when single-kernel weight 〉0.0064 g. The most effective model was based on a multilayer perceptron (MLP) neural network, consisting of 15 physical traits as variables, and a stability calculated as 99.4%. Germination percentage in a calibration set of seeds was 79.1% and the selection rate was 90.0%. These results indicated that the model was effective in predicting seed germination based on physical features and could be used as a guide for quality control in seed selection. Automated systems based on machine vision and model classifiers can contribute to reducing the costs and labor required in the selection of pepper seeds.
基金This research was done as part of TEKES-funded PanFlow project and as part of a project OPTIMI funded by the Academy of Finland (grant number 117587) in Micro- and Nanosystems Research Group, Tampere University of Technology, Finland.
文摘When characterizing flows in miniaturized channels, the determination of the dynamic contact angle is important. By measuring the dynamic contact angle, the flow properties of the flowing liquid and the effect of material properties on the flow can be characterized. A machine vision based system to measure the contact angle of front or rear menisci of a moving liquid plug is described in this article. In this research, transparent flow channels fabricated on thermoplastic polymer and sealed with an adhesive tape are used. The transparency of the channels enables image based monitoring and measurement of flow variables, including the dynamic contact angle. It is shown that the dynamic angle can be measured from a liquid flow in a channel using the image based measurement system. An image processing algorithm has been developed in a MATLAB environment. Images are taken using a CCD camera and the channels are illuminated using a custom made ring light. Two fitting methods, a circle and two parabolas, are experimented and the results are compared in the measurement of the dynamic contact angles.
基金supported by the National Natural Science Foundation of China (31471411)the Shanghai Agriculture Applied Technology Development Program,China ((2017)3-8-4)
文摘In this study, we established a dynamic morphological model using the accumulated thermal effectiveness and photosynthetic active radiation (PAR) (A-TEP), aiming to explore the relationship between muskmelon (Cucumis melo L.) fruit attributes and environmental factors. Muskmelon surface color was described by parameters of red, green, blue, hue, saturation and brightness (HSI). Three characteristic parameters, gray level co-occurrence matrix (GLCM), angular second moment (ASM), entropy, contrast, and the coverage rate were used to describe the process of muskmelon fruit netting formation. ASM was not significant difference during muskmelon fruit growth. The number and deep of netting stripes gradually increased with fruit growth. Coverage rate increased rapidly for 15-30 d after pollination. The vertical and horizontal diameters of muskmelon fruit were followed a logistic curve. And root mean squared errors (RMSE) between the simulated and measured vertical and horizontal diameters were 3.527 and 4.696 mm, respectively. RMSE of red, green, blue, saturation and brightness were 0.999, 2.690, 2.992, 0.033 and 5.51, respectively, and the RMSE for entropy, contrast and coverage rates were 0.077, 0.063 and 0.015, respectively, indicating a well consistent between measured and simulated values.
基金Supported by the Ministerial Level Advanced Research Foundation(51318020309)
文摘The objective of this study was to develop an online tool-wear-measurement scheme for small diameter end-mills based on machine vision to increase tool life and the production efficiency. The geometrical features of wear zone of each end mill were analyzed, and three tool wear criterions of small-diameter end mills were defined. With the uEye camera, macro lens and 3-axis micro milling machine, it was proved the feasibility of measuring flank wear with the milling tests on a 45# steel workpiece. The design of experiment (DOE) showed that Vc was the most remarkable effect factor for the flank wear of small-diameter end mill. The wear curve of the experiments of milling was very similar to the Taylor curve.
文摘Manual monitoring and seam tracking through watching weld pool images in real-time, by naked eyes or by industrial TV, are experience-depended, subjective, labor intensive, and sometimes biased. So it is necessary to realize the automation of computer-aided seam tracking. A PAW (plasma arc welding) seam tracking system was developed, which senses the molten pool and the seam in one frame by a vision sensor, and then detects the seam deviation to adjust the work piece motion adaptively to the seam position sensed by vision sensor. A novel molten pool area image-processing algorithm based on machine vision was proposed. The algorithm processes each image at the speed of 20 frames/second in real-time to extract three feature variables to get the seam deviation. It is proved experimentally that the algorithm is very fast and effective. Issues related to the algorithm are also discussed.