Lightweight Res-Connection Multi-Branch Network for Highly Accurate Crowd Counting and Localization

Crowd counting is a promising hotspot of computer vision involving crowd intelligence analysis,achieving tremendous success recently with the development of deep learning.However,there have been stillmany challenges including crowd multi-scale variations and high network complexity,etc.To tackle these issues,a lightweight Resconnection multi-branch network(LRMBNet)for highly accurate crowd counting and localization is proposed.Specifically,using improved ShuffleNet V2 as the backbone,a lightweight shallow extractor has been designed by employing the channel compression mechanism to reduce enormously the number of network parameters.A light multi-branch structure with different expansion rate convolutions is demonstrated to extract multi-scale features and enlarged receptive fields,where the information transmission and fusion of diverse scale features is enhanced via residual concatenation.In addition,a compound loss function is introduced for training themethod to improve global context information correlation.The proposed method is evaluated on the SHHA,SHHB,UCF-QNRF and UCF_CC_50 public datasets.The accuracy is better than those of many advanced approaches,while the number of parameters is smaller.The experimental results show that the proposed method achieves a good tradeoff between the complexity and accuracy of crowd counting,indicating a lightweight and high-precision method for crowd counting.

Deep Learning Based Efficient Crowd Counting System

Estimation of crowd count is becoming crucial nowadays,as it can help in security surveillance,crowd monitoring,and management for different events.It is challenging to determine the approximate crowd size from an image of the crowd’s density.Therefore in this research study,we proposed a multi-headed convolutional neural network architecture-based model for crowd counting,where we divided our proposed model into two main components:(i)the convolutional neural network,which extracts the feature across the whole image that is given to it as an input,and(ii)the multi-headed layers,which make it easier to evaluate density maps to estimate the number of people in the input image and determine their number in the crowd.We employed the available public benchmark crowd-counting datasets UCF CC 50 and ShanghaiTech parts A and B for model training and testing to validate the model’s performance.To analyze the results,we used two metrics Mean Absolute Error(MAE)and Mean Square Error(MSE),and compared the results of the proposed systems with the state-of-art models of crowd counting.The results show the superiority of the proposed system.

A multi-source information fusion layer counting method for penetration fuze based on TCN-LSTM

When employing penetration ammunition to strike multi-story buildings,the detection methods using acceleration sensors suffer from signal aliasing,while magnetic detection methods are susceptible to interference from ferromagnetic materials,thereby posing challenges in accurately determining the number of layers.To address this issue,this research proposes a layer counting method for penetration fuze that incorporates multi-source information fusion,utilizing both the temporal convolutional network(TCN)and the long short-term memory(LSTM)recurrent network.By leveraging the strengths of these two network structures,the method extracts temporal and high-dimensional features from the multi-source physical field during the penetration process,establishing a relationship between the multi-source physical field and the distance between the fuze and the target plate.A simulation model is developed to simulate the overload and magnetic field of a projectile penetrating multiple layers of target plates,capturing the multi-source physical field signals and their patterns during the penetration process.The analysis reveals that the proposed multi-source fusion layer counting method reduces errors by 60% and 50% compared to single overload layer counting and single magnetic anomaly signal layer counting,respectively.The model's predictive performance is evaluated under various operating conditions,including different ratios of added noise to random sample positions,penetration speeds,and spacing between target plates.The maximum errors in fuze penetration time predicted by the three modes are 0.08 ms,0.12 ms,and 0.16 ms,respectively,confirming the robustness of the proposed model.Moreover,the model's predictions indicate that the fitting degree for large interlayer spacings is superior to that for small interlayer spacings due to the influence of stress waves.

Dynamic time-correlated single-photon counting laser ranging

We demonstrate a photon counting laser ranging experiment with a four-channel single-photon detector(SPD). The multi-channel SPD improve the counting rate more than 4×10~7 cps, which makes possible for the distance measurement performed even in daylight. However, the time-correlated single-photon counting(TCSPC) technique cannot distill the signal easily while the fast moving targets are submersed in the strong background. We propose a dynamic TCSPC method for fast moving targets measurement by varying coincidence window in real time. In the experiment, we prove that targets with velocity of 5 km/s can be detected according to the method, while the echo rate is 20% with the background counts of more than 1.2×10~7 cps.

Improved transient electroluminescence technique based on time-correlated single-photon counting technology to evaluate organic mobility

The transient electroluminescence(EL)technique is widely used to evaluate the carrier mobility in the field of organic light emitting diodes.The traditional analog detection strategy using oscilloscopes is generally limited since the background noise causes an underestimation of the mobility value.In this paper,we utilize time-correlated single-photon counting(TCSPC)to probe the transient EL for mobility calculation.The measurements on tris(8-hydroxyquinoline)aluminum(Alq3)show that the electron mobilities obtained using the TCSPC technique are slightly higher than those obtained from the analog method at all the investigated voltages.Moreover,the TCSPC mobilities demonstrate weaker dependence on the root of electrical field compared to the oscilloscope mobilities.These improvements are attributed to the unique principle of TCSPC,which quantifies the EL intensity by counting the number of single-photon pulses,improving its single-photon sensitivity and elimi-nating the negative impacts of electrical noise.These advantages make TCSPC a powerful technique in the characterization of time-resolved electroluminescence.

Counting of alpha particle tracks on imaging plate based on a convolutional neural network

Imaging plates are widely used to detect alpha particles to track information,and the number of alpha particle tracks is affected by the overlapping and fading effects of the track information.In this study,an experiment and a simulation were used to calibrate the efficiency parameter of an imaging plate,which was used to calculate the grayscale.Images were created by using grayscale,which trained the convolutional neural network to count the alpha tracks.The results demonstrated that the trained convolutional neural network can evaluate the alpha track counts based on the source and background images with a wider linear range,which was unaffected by the overlapping effect.The alpha track counts were unaffected by the fading effect within 60 min,where the calibrated formula for the fading effect was analyzed for 132.7 min.The detection efficiency of the trained convolutional neural network for inhomogeneous ^(241)Am sources(2π emission)was 0.6050±0.0399,whereas the efficiency curve of the photo-stimulated luminescence method was lower than that of the trained convolutional neural network.

基于Count Sketch的预处理贪婪Kaczmarz方法

在贪婪Kaczmarz方法中,通过对系数矩阵进行正交三角分解引入右预处理子能够提高贪婪Kaczmarz方法的收敛速率。但在系数矩阵的行数远大于列数的情况下,正交三角分解的成本过高。为降低预处理的成本,通过引入Count Sketch变换,提出了基于Count Sketch的预处理贪婪Kaczmarz方法,并对新方法进行了收敛性分析。理论分析说明了新方法在系数矩阵条件数较大时比已有方法具有更好的收敛速率。数值实验验证了新方法的有效性。

Photon-counting chirped amplitude modulation lidar system using superconducting nanowire single-photon detector at 1550-nm wavelength

We demonstrate a photon-counting chirped amplitude modulation （CAM） light detection and ranging （lidar） system incorporating a superconducting nanowire single-photon detector （SNSPD） and operated at a wavelength of 1550 nm. The distance accuracy of the lidar system was determined by the CAM bandwidth and signal-to-noise ratio （SNR） of an intermediate frequency （IF） signal. Owing to a short dead time （10 ns） and negligible dark count rate （70 Hz） of the SNSPD, the obtained IF signal attained an SNR of 42 dB and the direct distance accuracy was improved to 3 mm when the modulation bandwidth of the CAM signal was 240 MHz and the modulation period was 1 ms.

RPNet: Rice plant counting after tillering stage based on plant attention and multiple supervision network

Rice is a major food crop and is planted worldwide. Climatic deterioration, population growth, farmland shrinkage, and other factors have necessitated the application of cutting-edge technology to achieve accurate and efficient rice production. In this study, we mainly focus on the precise counting of rice plants in paddy field and design a novel deep learning network, RPNet, consisting of four modules: feature encoder, attention block, initial density map generator, and attention map generator. Additionally, we propose a novel loss function called RPloss. This loss function considers the magnitude relationship between different sub-loss functions and ensures the validity of the designed network. To verify the proposed method, we conducted experiments on our recently presented URC dataset, which is an unmanned aerial vehicle dataset that is quite challenged at counting rice plants. For experimental comparison, we chose some popular or recently proposed counting methods, namely MCNN, CSRNet, SANet, TasselNetV2, and FIDTM. In the experiment, the mean absolute error(MAE), root mean squared error(RMSE), relative MAE(rMAE) and relative RMSE(rRMSE) of the proposed RPNet were 8.3, 11.2, 1.2% and 1.6%, respectively,for the URC dataset. RPNet surpasses state-of-the-art methods in plant counting. To verify the universality of the proposed method, we conducted experiments on the well-know MTC and WED datasets. The final results on these datasets showed that our network achieved the best results compared with excellent previous approaches. The experiments showed that the proposed RPNet can be utilized to count rice plants in paddy fields and replace traditional methods.

Lightweight Fish Bait Particle Counting Method Based on Pruning and Shift Quantization

In the process of aquaculture,monitoring the number of fish bait particles is of great significance to improve the growth and welfare of fish.Although the counting method based on onvolutional neural network(CNN)achieve good accuracy and applicability,it has a high amount of parameters and computation,which limit the deployment on resource-constrained hardware devices.In order to solve the above problems,this paper proposes a lightweight bait particle counting method based on shift quantization and model pruning strategies.Firstly,we take corresponding lightweight strategies for different layers to flexibly balance the counting accuracy and performance of the model.In order to deeply lighten the counting model,the redundant and less informative weights of the model are removed through the combination of model quantization and pruning.The experimental results show that the compression rate is nearly 9 times.Finally,the quantization candidate value is refined by introducing a power-of-two addition term,which improves the matches of the weight distribution.By analyzing the experimental results,the counting loss at 3 bit is reduced by 35.31%.In summary,the lightweight bait particle counting model proposed in this paper achieves lossless counting accuracy and reduces the storage and computational overhead required for running convolutional neural networks.

A new software for automated counting of glistenings in intraocular lenses in vivo

AIM:To assess the performance of a bespoke software for automated counting of intraocular lens(IOL)glistenings in slit-lamp images.METHODS:IOL glistenings from slit-lamp-derived digital images were counted manually and automatically by the bespoke software.The images of one randomly selected eye from each of 34 participants were used as a training set to determine the threshold setting that gave the best agreement between manual and automatic grading.A second set of 63 images,selected using randomised stratified sampling from 290 images,were used for software validation.The images were obtained using a previously described protocol.Software-derived automated glistenings counts were compared to manual counts produced by three ophthalmologists.RESULTS:A threshold value of 140 was determined that minimised the total deviation in the number of glistenings for the 34 images in the training set.Using this threshold value,only slight agreement was found between automated software counts and manual expert counts for the validating set of 63 images(κ=0.104,95%CI,0.040-0.168).Ten images(15.9%)had glistenings counts that agreed between the software and manual counting.There were 49 images(77.8%)where the software overestimated the number of glistenings.CONCLUSION:The low levels of agreement show between an initial release of software used to automatically count glistenings in in vivo slit-lamp images and manual counting indicates that this is a non-trivial application.Iterative improvement involving a dialogue between software developers and experienced ophthalmologists is required to optimise agreement.The results suggest that validation of software is necessary for studies involving semi-automatic evaluation of glistenings.

A Computer Vision-Based System for Metal Sheet Pick Counting

Inventory counting is crucial to manufacturing industries in terms of inventory management,production,and procurement planning.Many companies currently require workers to manually count and track the status of materials,which are repetitive and non-value-added activities but incur significant costs to the companies as well as mental fatigue to the employees.This research aims to develop a computer vision system that can automate the material counting activity without applying any marker on the material.The type of material of interest is metal sheet,whose shape is simple,a large rectangular shape,yet difficult to detect.The use of computer vision technology can reduce the costs incurred fromthe loss of high-value materials,eliminate repetitive work requirements for skilled labor,and reduce human error.A computer vision system is proposed and tested on a metal sheet picking process formultiple metal sheet stacks in the storage area by using one video camera.Our results show that the proposed computer vision system can count the metal sheet picks under a real situation with a precision of 97.83%and a recall of 100%.

A Deep Learning-Based Crowd Counting Method and System Implementation on Neural Processing Unit Platform

In this paper, a deep learning-based method is proposed for crowdcountingproblems. Specifically, by utilizing the convolution kernel densitymap, the ground truth is generated dynamically to enhance the featureextractingability of the generator model. Meanwhile, the “cross stage partial”module is integrated into congested scene recognition network (CSRNet) toobtain a lightweight network model. In addition, to compensate for the accuracydrop owing to the lightweight model, we take advantage of “structuredknowledge transfer” to train the model in an end-to-end manner. It aimsto accelerate the fitting speed and enhance the learning ability of the studentmodel. The crowd-counting system solution for edge computing is alsoproposed and implemented on an embedded device equipped with a neuralprocessing unit. Simulations demonstrate the performance improvement ofthe proposed solution in terms of model size, processing speed and accuracy.The performance on the Venice dataset shows that the mean absolute error(MAE) and the root mean squared error (RMSE) of our model drop by32.63% and 39.18% compared with CSRNet. Meanwhile, the performance onthe ShanghaiTech PartB dataset reveals that the MAE and the RMSE of ourmodel are close to those of CSRNet. Therefore, we provide a novel embeddedplatform system scheme for public safety pre-warning applications.

Improvement of Counting Sorting Algorithm

By analyzing the internal features of counting sorting algorithm. Two improvements of counting sorting algorithms are proposed, which have a wide range of applications and better efficiency than the original counting sort while maintaining the original stability. Compared with the original counting sort, it has a wider scope of application and better time and space efficiency. In addition, the accuracy of the above conclusions can be proved by a large amount of experimental data.

Robust Counting in Overcrowded Scenes Using Batch-Free Normalized Deep ConvNet

The analysis of overcrowded areas is essential for flow monitoring,assembly control,and security.Crowd counting’s primary goal is to calculate the population in a given region,which requires real-time analysis of congested scenes for prompt reactionary actions.The crowd is always unexpected,and the benchmarked available datasets have a lot of variation,which limits the trained models’performance on unseen test data.In this paper,we proposed an end-to-end deep neural network that takes an input image and generates a density map of a crowd scene.The proposed model consists of encoder and decoder networks comprising batch-free normalization layers known as evolving normalization(EvoNorm).This allows our network to be generalized for unseen data because EvoNorm is not using statistics from the training samples.The decoder network uses dilated 2D convolutional layers to provide large receptive fields and fewer parameters,which enables real-time processing and solves the density drift problem due to its large receptive field.Five benchmark datasets are used in this study to assess the proposed model,resulting in the conclusion that it outperforms conventional models.

Research on the correlation between the dual diffusion behavior of zinc in InGaAs/InP single-photon avalanche photodiodes and device performance

The development of InGaAs/InP single-photon avalanche photodiodes(SPADs)necessitates the utiliza-tion of a two-element diffusion technique to achieve accurate manipulation of the multiplication width and the dis-tribution of its electric field.Regarding the issue of accurately predicting the depth of diffusion in InGaAs/InP SPAD,simulation analysis and device development were carried out,focusing on the dual diffusion behavior of zinc atoms.A formula of X_(j)=k√t-t_(0)+c to quantitatively predict the diffusion depth is obtained by fitting the simulated twice-diffusion depths based on a two-dimensional(2D)model.The 2D impurity morphologies and the one-dimensional impurity profiles for the dual-diffused region are characterized by using scanning electron micros-copy and secondary ion mass spectrometry as a function of the diffusion depth,respectively.InGaAs/InP SPAD devices with different dual-diffusion conditions are also fabricated,which show breakdown behaviors well consis-tent with the simulated results under the same junction geometries.The dark count rate(DCR)of the device de-creased as the multiplication width increased,as indicated by the results.DCRs of 2×10^(6),1×10^(5),4×10^(4),and 2×10^(4) were achieved at temperatures of 300 K,273 K,263 K,and 253 K,respectively,with a bias voltage of 3 V,when the multiplication width was 1.5µm.These results demonstrate an effective prediction route for accu-rately controlling the dual-diffused zinc junction geometry in InP-based planar device processing.

Improved decoy-state quantum key distribution with uncharacterized heralded single-photon sources

Encoding system plays a significant role in quantum key distribution(QKD).However,the security and performance of QKD systems can be compromised by encoding misalignment due to the inevitable defects in realistic devices.To alleviate the influence of misalignments,a method exploiting statistics from mismatched basis is proposed to enable uncharacterized sources to generate secure keys in QKD.In this work,we propose a scheme on four-intensity decoy-state quantum key distribution with uncharacterized heralded single-photon sources.It only requires the source states are prepared in a two-dimensional Hilbert space,and can thus reduce the complexity of practical realizations.Moreover,we carry out corresponding numerical simulations and demonstrate that our present four-intensity decoy-state scheme can achieve a much higher key rate compared than a three-intensity decoy-state method,and meantime it can obtain a longer transmission distance compared than the one using weak coherent sources.

The intensity detection of single-photon detectors based on photon counting probability density statistics

Single-photon detectors possess the ultra-high sensitivity, but they cannot directly respond to signal intensity. Conven- tional methods adopt sampling gates with fixed width and count the triggered number of sampling gates, which is capable of obtaining photon counting probability to estimate the echo signal intensity. In this paper, we not only count the number of triggered sampling gates, but also record the triggered time position of photon counting pulses. The photon counting probability density distribution is obtained through the statistics of a series of the triggered time positions. Then Minimum Variance Unbiased Estimation （MVUE） method is used to estimate the echo signal intensity. Compared with conventional methods, this method can improve the estimation accuracy of echo signal intensity due to the acquisition of more detected information. Finally, a proof-of-principle laboratory system is established. The estimation accuracy of echo signal intensity is discussed and a high accuracy intensity image is acquired under low-light level environments.

Rapid Detection of Somatic Cell Count Based on Hybrid Variable Selection Method

Somatic cell count detection is the daily work of dairy farms to monitor the health of cows.The feasibility of applying near-infrared spectroscopy to somatic cell count detection was researched in this paper.Milk samples with different somatic cell counts were collected and preprocessing methods were studied.Variable selection algorithm based on hybrid strategy and modelling method based on ensemble learning were explored for somatic cell count detection.Detection model was used to diagnose subclinical mastitis and the results showed that near-infrared spectroscopy could be a tool to realize rapid detection of somatic cell count in milk.

Platelet counts to spleen diameter ratio:A promising noninvasive tool for predicting esophageal varices in cirrhosis patients

BACKGROUND Liver cirrhosis is the end stage of progressive liver fibrosis as a consequence of chronic liver inflammation,wherein the standard hepatic architecture is replaced by regenerative hepatic nodules,which eventually lead to liver failure.Cirrhosis without any symptoms is referred to as compensated cirrhosis.Complications such as ascites,variceal bleeding,and hepatic encephalopathy indicate the onset of decompensated cirrhosis.Gastroesophageal varices are the hallmark of clini-cally significant portal hypertension.AIM To determine the accuracy of the platelet count-to-spleen diameter(PC/SD)ratio to evaluate esophageal varices(EV)in patients with cirrhosis.METHODS This retrospective observational study was conducted at Tikur Anbessa Specia-lized Hospital and Adera Medical Center from January 1,2019,to December 30,2023.Data were collected via chart review and direct patient interviews using structured questionnaires.The data were exported to the SPSS software version 26 for analysis and clearance.A receiver operating characteristic curve was plotted for splenic diameter,platelet count,and PC/SD ratio to obtain sensitivity,speci-ficity,positive predictive value,negative predictive value,positive likelihood ratio,and negative likelihood ratio.RESULTS Of the 140 participants,67%were men.Hepatitis B(38%)was the most common cause of cirrhosis,followed by cryptogenic cirrhosis(28%)and hepatitis C(16%).Approximately 83.6%of the participants had endoscopic evidence of EV,whereas 51.1%had gastric varices.Decompensated cirrhosis and PC were associated with the presence of EV with adjusted odds ratios of 12.63(95%CI:3.16-67.58,P=0.001)and 0.14(95%CI:0.037-0.52,P=0.004),respectively.A PC/SD ratio<1119 had a sensitivity of 86.32%and specificity of 70%with area under the curve of 0.835(95%CI:0.736-0.934,P<0.001).CONCLUSION A PC/SD ratio<1119 predicts EV in patients with cirrhosis.It is a valuable,noninvasive tool for EV risk assess-ment in resource-limited settings.