Techniques of Image Processing Based on Artificial Neural Networks

This paper presented an online quality inspection system based on artificial neural networks. Chromatism classification and edge detection are two difficult problems in glass steel surface quality inspection. Two artificial neural networks were made and the two problems were solved. The one solved chromatism classification. Hue, saturation and their probability of three colors, whose appearing probabilities were maximum in color histogram, were selected as input parameters, and the number of output node could be adjusted with the change of requirement. The other solved edge detection. In this neutral network, edge detection of gray scale image was able to be tested with trained neural networks for a binary image. It prevent the difficulty that the number of needed training samples was too large if gray scale images were directly regarded as training samples. This system is able to be applied to not only glass steel fault inspection but also other product online quality inspection and classification.

Edge detection of potential field data based on image processing methods

The conventional methods of edge detection can roughly delineate edge position of geological bodies,but there are still some problems such as low detection accuracy and being susceptible to noise interference.In this paper,three image processing methods,Canny,Lo G and Sobel operators are briefly introduced,and applied to edge detection to determine the edge of geological bodies.Furthermore,model data is built to analyze the edge detection ability of this image processing methods,and compare with conventional methods.Combined with gravity anomaly of Sichuan basin and magnetic anomaly of Zhurihe area,the detection effect of image processing methods is further verified in real data.The results show that image processing methods can be applied to effectively identify the edge of geological bodies.Moreover,when both positive and negative anomalies exist and noise is abundant,fake edge can be avoided and edge division is clearer,and satisfactory results of edge detection are obtained.

Design of multilayer cellular neural network based on memristor crossbar and its application to edge detection

Memristor with memory properties can be applied to connection points(synapses)between cells in a cellular neural network(CNN).This paper highlights memristor crossbar-based multilayer CNN(MCM-CNN)and its application to edge detection.An MCM-CNN is designed by adopting a memristor crossbar composed of a pair of memristors.MCM-CNN based on the memristor crossbar with changeable weight is suitable for edge detection of a binary image and a color image considering its characteristics of programmablization and compactation.Figure of merit(FOM)is introduced to evaluate the proposed structure and several traditional edge detection operators for edge detection results.Experiment results show that the FOM of MCM-CNN is three times more than that of the traditional edge detection operators.

Feature fusion method for edge detection of color images

A novel feature fusion method is proposed for the edge detection of color images. Except for the typical features used in edge detection, the color contrast similarity and the orientation consistency are also selected as the features. The four features are combined together as a parameter to detect the edges of color images. Experimental results show that the method can inhibit noisy edges and facilitate the detection for weak edges. It has a better performance than conventional methods in noisy environments.

SAR image despeckling based on edge detection and nonsubsampled second generation bandelets

To preserve the sharp features and details of the synthetic aperture radar （SAR） image effectively when despeckling, a despeckling algorithm with edge detection in nonsubsampled second generation bandelet transform （NSBT） domain is proposed. First, the Canny operator is utilized to detect and remove edges from the SAR image. Then the NSBT which has an optimal approximation to the edges of images and a hard thresholding rule are used to approximate the details while despeckling the edge-removed image. Finally, the removed edges are added to the reconstructed image. As the edges axe detected and protected, and the NSBT is used, the proposed algorithm reaches the state-of-the-art effect which realizes both despeckling and preserving edges and details simultaneously. Experimental results show that both the subjective visual effect and the mainly objective performance indexes of the proposed algorithm outperform that of both Bayesian wavelet shrinkage with edge detection and Bayesian least square-Gaussian scale mixture （BLS-GSM）.

Real-time image processing and display in object size detection based on VC++

Real-time detection for object size has now become a hot topic in the testing field and image processing is the core algorithm. This paper focuses on the processing and display of the collected dynamic images to achieve a real-time image pro- cessing for the moving objects. Firstly, the median filtering, gain calibration, image segmentation, image binarization, cor- ner detection and edge fitting are employed to process the images of the moving objects to make the image close to the real object. Then, the processed images are simultaneously displayed on a real-time basis to make it easier to analyze, understand and identify them, and thus it reduces the computation complexity. Finally, human-computer interaction （HCI）-friendly in- terface based on VC ＋＋ is designed to accomplish the digital logic transform, image processing and real-time display of the objects. The experiment shows that the proposed algorithm and software design have better real-time performance and accu- racy which can meet the industrial needs.

An Improved Algorithm for Image Edge Detection Based on Lifting Scheme

Wavelet transform is an ideal way for edge detection because of its multi-scale property, localization both in time and frequency domain, sensitivity to the abrupt change of signals, and so on. An improved algorithm for image edge detection based on Lifting Scheme is proposed in this paper. The simulation results show that our improved method can better reflect edge information of images.

Edge Detection Based on Generative Adversarial Networks

Aiming at the problem that the detection effect of traditional edge detection algorithm is not good,and the problem that the existing edge detection algorithm based on convolution network cannot solve the thick edge problem from the model itself,this paper proposes a new edge detection method based on the generative adversarial network.The confrontation network consists of generator network and discriminator network,generator network is composed of U-net network and discriminator network is composed of five-layer convolution network.In this paper,we use BSDS500 training data set to train the model.Finally,several images are randomly selected from BSDS500 test set to compare with the results of traditional edge detection algorithm and HED algorithm.The results of BSDS500 benchmark test show that the ODS and OIS indices of the proposed method are 0.779 and 0.782 respectively,which are much higher than those of traditional edge detection algorithms,and the indices of HED algorithm using non-maximum suppression are similar.

An Optimized and Hybrid Framework for Image Processing Based Network Intrusion Detection System

The network infrastructure has evolved rapidly due to the everincreasing volume of users and data.The massive number of online devices and users has forced the network to transform and facilitate the operational necessities of consumers.Among these necessities,network security is of prime significance.Network intrusion detection systems(NIDS)are among the most suitable approaches to detect anomalies and assaults on a network.However,keeping up with the network security requirements is quite challenging due to the constant mutation in attack patterns by the intruders.This paper presents an effective and prevalent framework for NIDS by merging image processing with convolution neural networks(CNN).The proposed framework first converts non-image data from network traffic into images and then further enhances those images by using the Gabor filter.The images are then classified using a CNN classifier.To assess the efficacy of the recommended method,four benchmark datasets i.e.,CSE-CIC-IDS2018,CIC-IDS-2017,ISCX-IDS 2012,and NSL-KDD were used.The proposed approach showed higher precision in contrast with the recent work on the mentioned datasets.Further,the proposed method is compared with the recent well-known image processing methods for NIDS.

Image Edge Detection Based on Oscillation

A new method for image edge detection based on a pulse neural network is proposed in this paper. The network is locally connected. The external input of each neuron of the network is gray value of the corresponding pixel. The synchrony of the neuron and its neighbors is detected by detection neurons. The edge of the image can be read off at minima of the total activity of the detection neurons.

Automatic recognition of defects in plasma-facing material using image processing technology

Observing and analyzing surface images is critical for studying the interaction between plasma and irradiated plasma-facing materials.This paper presents a method for the automatic recognition of bubbles in transmission electron microscope(TEM)images of W nanofibers using image processing techniques and convolutional neural network(CNN).We employ a three-stage approach consisting of Otsu,local-threshold,and watershed segmentation to extract bubbles from noisy images.To address over-segmentation,we propose a combination of area factor and radial pixel intensity scanning.A CNN is used to recognize bubbles,outperforming traditional neural network models such as Alex Net and Google Net with an accuracy of 97.1%and recall of 98.6%.Our method is tested on both clear and blurred TEM images,and demonstrates humanlike performance in recognizing bubbles.This work contributes to the development of quantitative image analysis in the field of plasma-material interactions,offering a scalable solution for analyzing material defects.Overall,this study's findings establish the potential for automatic defect recognition and its applications in the assessment of plasma-material interactions.This method can be employed in a variety of specialties,including plasma physics and materials science.

AN ALGORITHM OF EDGE DETECTION BASED ON ENTROPY OPERATOR

This paper presents an algorithm of edge detection in image processing. A new entropy operator and threshold estimation technique are effectively proposed. The algorithm overcomes some drawbacks of Shiozaki operator. It not only has higher speed but also can extract the edge better. Finally, an example of 2D image is given to demonstrate the usefulness and advantages of the algorithm.

Prediction of Pitting Corrosion Mass Loss for 304 Stainless Steel by Image Processing and BP Neural Network

Image processing technique was employed to analyze pitting corrosion morphologies of 304 stainless steel exposed to FeCl3 environments. BP neural network models were developed for the prediction of pitting corrosion mass loss using the obtained data of the total and the average pit areas which were extracted from pitting binary image. The results showed that the predicted results obtained by the 2-5-1 type BP neural network model are in good agreement with the experimental data of pitting corrosion mass loss. The maximum relative error of prediction is 6.78%.

Numerical simulation of direct shear tests on mechanical properties of talus deposits based on self-adaptive PCNN digital image processing

The macro mechanical properties of materials with characteristics of large scale and complicated structural composition can be analyzed through its reconstructed meso-structures.In this work,the meso-structures of talus deposits that widely exist in the hydro-power engineering in the southwest of China were first reconstructed by small particles according to the in-situ photographs based on the self-adaptive PCNN digital image processing,and then numerical direct shear tests were carried out for studying the mechanical properties of talus deposits.Results indicate that the reconstructed meso-structures of talus deposits are more consistent with the actual situation because the self-adaptive PCNN digital image processing has a higher discrimination in the details of soil-rock segmentation.The existence and random distribution of rock blocks make the initial shear stiffness,the peak strength and the residual strength higher than those of the "pure soil" with particle size less than 1.25 cm apparently,but reduce the displacements required for the talus deposits reaching its peak shear strength.The increase of rock proportion causes a significant improvement in the internal friction angle of talus deposit,which to a certain degree leads to the characteristics of shear stress-displacement curves having a changing trend from the plastic strain softening deformation to the nonlinear strain hardening deformation,while an unconspicuous increase in cohesion.The uncertainty and heterogeneity of rock distributions cause the differences of rock proportion within shear zone,leading to a relatively strong fluctuation in peak strengths during the shear process,while movement features of rock blocks,such as translation,rotation and crossing,expand the scope of shear zone,increase the required shear force,and also directly lead to the misjudgment that the lower shear strength is obtained from the samples with high rock proportion.That,however,just explains the reason why the shear strength gained from a small amount of indoor test data is not consistent with engineering practice.

Detection of “Swollen Shoot” Disease in Ivorian Cocoa Trees via Convolutional Neural Networks

Recent advances in diagnostics have made image analysis one of the main areas of research and development. Selecting and calculating these characteristics of a disease is a difficult task. Among deep learning techniques, deep convolutional neural networks are actively used for image analysis. This includes areas of application such as segmentation, anomaly detection, disease classification, computer-aided diagnosis. The objective which we aim in this article is to extract information in an effective way for a better diagnosis of the plants attending the disease of “swollen shoot”.

Novel welding image processing method based on fractal theory

Computer vision has come into used in the fields of welding process control and automation. In order to improve precision and rapidity of welding image processing, a novel method based on fractal theory has been put forward in this paper. Compared with traditional methods, the image is preliminarily processed in the macroscopic regions then thoroughly analyzed in the microscopic regions in the new method. With which, an image is divided up to some regions according to the different fractal characters of image edge, and the fuzzy regions including image edges are detected out, then image edges are identified with Sobel operator and curved by LSM (Lease Square Method). Since the data to be processed have been decreased and the noise of image has been reduced, it has been testified through experiments that edges of weld seam or weld pool could be recognized correctly and quickly.

Application of Neural Networks to Matlab Analyzed Hyperspectral Images for Characterization of Composite Structures

A novel approach to damage detection in composite structures using hyperspectral image index analysis algorithm with neural network modeling employing Weight Elimination Algorithm (WEA) is presented and discussed. The matrix band based technique allows the monitoring and analysis of a component’s structure based on correlation between sequentially pulsed thermal images. The technique produces several matrices resulting from frame deviation and pixel redistribution calculations with ability for prediction. The obtained results proved the technique to be capable of identifying damaged components with ability to model various types of damage under different conditions.

Negation Scope Detection with Recurrent Neural Networks Models in Review Texts

Identifying negation scopes in a text is an important subtask of information extraction, that can benefit other natural language processing tasks, like relation extraction, question answering and sentiment analysis. And serves the task of social media text understanding. The task of negation scope detection can be regarded as a token-level sequence labeling problem. In this paper, we propose different models based on recurrent neural networks (RNNs) and word embedding that can be successfully applied to such tasks without any task-specific feature engineering efforts. Our experimental results show that RNNs, without using any hand-crafted features, outperform feature-rich CRF-based model.

Detection of Oscillations in Process Control Loops From Visual Image Space Using Deep Convolutional Networks

Oscillation detection has been a hot research topic in industries due to the high incidence of oscillation loops and their negative impact on plant profitability.Although numerous automatic detection techniques have been proposed,most of them can only address part of the practical difficulties.An oscillation is heuristically defined as a visually apparent periodic variation.However,manual visual inspection is labor-intensive and prone to missed detection.Convolutional neural networks(CNNs),inspired by animal visual systems,have been raised with powerful feature extraction capabilities.In this work,an exploration of the typical CNN models for visual oscillation detection is performed.Specifically,we tested MobileNet-V1,ShuffleNet-V2,Efficient Net-B0,and GhostNet models,and found that such a visual framework is well-suited for oscillation detection.The feasibility and validity of this framework are verified utilizing extensive numerical and industrial cases.Compared with state-of-theart oscillation detectors,the suggested framework is more straightforward and more robust to noise and mean-nonstationarity.In addition,this framework generalizes well and is capable of handling features that are not present in the training data,such as multiple oscillations and outliers.

Intelligent Prediction Approach for Diabetic Retinopathy Using Deep Learning Based Convolutional Neural Networks Algorithm by Means of Retina Photographs

Retinopathy is a human eye disease that causes changes in retinal blood vessels that leads to bleed,leak fluid and vision impairment.Symptoms of retinopathy are blurred vision,changes in color perception,red spots,and eye pain and it cannot be detected with a naked eye.In this paper,a new methodology based on Convolutional Neural Networks(CNN)is developed and proposed to intelligent retinopathy prediction and give a decision about the presence of retinopathy with automatic diabetic retinopathy screening with accurate diagnoses.The CNN model is trained by different images of eyes that have retinopathy and those which do not have retinopathy.The fully connected layers perform the classification process of the images from the dataset with the pooling layers minimize the coherence among the adjacent layers.The feature loss factor increases the label value to identify the patterns with the kernel-based matching.The performance of the proposed model is compared with the related methods of DREAM,KNN,GD-CNN and SVM.Experimental results show that the proposed CNN performs better.