Atrous convolutions spatial pyramid network for crowd counting and density estimation

Segmentation of void defects in X-ray images of chip solder joints based on PCB-DeepLabV3 algorithm

Defects within chip solder joints are usually inspected visually for defects using X-ray imaging to obtain images. The phenomenon of voids inside solder joints is one of the most likely types of defects in the soldering process, and accurate detection of voids becomes difficult due to their irregular shapes, varying sizes, and defocused edges. To address this problem, an X-ray void image segmentation algorithm based on improved PCB-DeepLabV3 is proposed. Firstly, to meet the demand for lightweight and easy deployment in industrial scenarios, mobilenetv2 is used as the feature extraction backbone network of the PCB-DeepLabV3 model; then, Attentional multi-scale two-space pyramid pooling network (AMTPNet) is designed to optimize the shallow feature edges and to improve the ability to capture detailed information; finally, image cropping and cleaning methods are designed to enhance the training dataset, and the improved PCB-DeepLabV3 is applied to the training dataset. The improved PCB-DeepLabV3 model is used to segment the void regions within the solder joints and compared with the classical semantic segmentation models such as Unet, SegNet, PSPNet, and DeeplabV3. The proposed new method enables the solder joint void inspection to get rid of the traditional way of visual inspection, realize intelligent upgrading, and effectively improve the problem of difficult segmentation of the target virtual edges, to obtain the inspection results with higher accuracy.

Foreground Segmentation-Based Density Grading Networks for Crowd Counting

Estimating object counts within a single image or video frame represents a challenging yet pivotal task in the field of computer vision. Its increasing significance arises from its versatile applications across various domains, including public safety and urban planning. Among the various object counting tasks, crowd counting is particularly notable for its critical role in social security and urban planning. However, intricate backgrounds in images often lead to misidentifications, wherein the complex background is mistaken as the foreground, thereby inflating forecasting errors. Additionally, the uneven distribution of crowd density within the foreground further exacerbates predictive errors of the network. This paper introduces a novel architecture with a three-branch structure aimed at synergistically incorporating hierarchical foreground information and global scale information into density map estimation, thereby achieving more precise counting results. Hierarchical foreground information guides the network to perform distinct operations on regions with varying densities, while global scale information evaluates the overall density level of the image and adjusts the model's global predictions accordingly. We also systematically investigate and compare three potential locations for integrating hierarchical foreground information into the density estimation network, ultimately determining the most effective placement.Through extensive comparative experiments across three datasets, we demonstrate the superior performance of our proposed method.

Generative Adversarial Networks to Improve the Robustness of Visual Defect Segmentation by Semantic Networks in Manufacturing Components

This paper describes the application of Semantic Networks for the detection of defects in images of metallic manufactured components in a situation where the number of available samples of defects is small, which is rather common in real practical environments. In order to overcome this shortage of data, the common approach is to use conventional data augmentation techniques. We resort to Generative Adversarial Networks (GANs) that have shown the capability to generate highly convincing samples of a specific class as a result of a game between a discriminator and a generator module. Here, we apply the GANs to generate samples of images of metallic manufactured components with specific defects, in order to improve training of Semantic Networks (specifically DeepLabV3+ and Pyramid Attention Network (PAN) networks) carrying out the defect detection and segmentation. Our process carries out the generation of defect images using the StyleGAN2 with the DiffAugment method, followed by a conventional data augmentation over the entire enriched dataset, achieving a large balanced dataset that allows robust training of the Semantic Network. We demonstrate the approach on a private dataset generated for an industrial client, where images are captured by an ad-hoc photometric-stereo image acquisition system, and a public dataset, the Northeastern University surface defect database (NEU). The proposed approach achieves an improvement of 7% and 6% in an intersection over union (IoU) measure of detection performance on each dataset over the conventional data augmentation.

Intelligent Measurement of Morphological Characteristics of Fish Using Improved U-Net

In the smart mariculture, batch testing of breeding traits is a key issue in the breeding of improved fish varieties. The body length (BL), body width (BW) and body area (BA) features of fish are important indicators. They are of great significance in breeding, feeding and classification. To accurately and intelligently obtain the morphological characteristic sizes of fish in actual scenes, data augmentation is first used to greatly expand the published fish dataset, thereby ensuring the robustness of the training model. Then, an improved U-net segmentation and measurement algorithm is proposed, which uses a dilated convolution with a dilation rate 2 and a convolution to partially replace the convolution in the original U-net. This operation can enlarge the partial convolution receptive field and achieve more accurate segmentation for large targets in the scene. Finally, a line fitting method based on the least squares method is proposed, which is combined with the body shape features of fish and can accurately measure the BL and BW of inclined fish. Experimental results show that the Mean Intersection over Union (mIoU) is 97.6% and the average relative error of the area is 0.69%. Compared with the unimproved U-net, the average relative error of the area is reduced to about half. Moreover, with the improved U-net and the line fitting method, the average relative error of BL and the average relative error of BW of inclined fish decrease to 0.37% and 0.61%, respectively.

Fault Diagnosis of Rolling Bearings Based on a Residual Dilated Pyramid Network and Full Convolutional Denoising Autoencoder

Intelligent fault diagnosis algorithm for rolling bearings has received increasing attention. However, in actual industrial environments, most rolling bearings work under severe working conditions of variable speed and strong noise, which makes the performance of many intelligent fault diagnosis methods deteriorate sharply. In this regard, this paper proposes a new intelligent diagnosis algorithm for rolling bearing faults based on a residual dilated pyramid network and full convolutional denoising autoencoder (RDPN-FCDAE). First, a continuous wavelet transform (CWT) is used to convert original vibration signals into time-frequency images. Secondly, a deep two-stage RDPN-FCDAE model is constructed, which is divided into three parts: encoding network, decoding network and classification network. In order to obtain efficient expression of data denoising feature of encoding network, time-frequency images are first input into the encoding-decoding network for unsupervised pre-training. Then pre-trained coding network and classification network are combined into residual dilated pyramid full convolutional network (RDPFCN) for parameter fine-tuning and testing. The proposed method is applied to bearing vibration datasets of test rig with different speeds and noise modes. Compared with representative machine learning and deep learning method, the results show that the algorithm proposed is superior to other methods in diagnostic accuracy, noise robustness and feature segmentation ability.

Indoor Crowd 3D Localization in Big Buildings from Wi-Fi Access Anonymous Data

Indoor crowd localization and counting in big public buildings pose problems of infrastructure deployment, signal processing, and privacy. Conventional approaches based on optical cameras, either in the visible or infrared range, received signal strength in wireless networks, sound or chemical sensing in sensor networks need careful calibration, noise removal, and sophisticated data processing to achieve results in limited scenarios. Moreover, personal data protection is a growing concern, so that detection methods that preserve the privacy of people are highly desirable. The aim of this paper is to provide a technique that may generate estimations of the localization of people in a big public building using anonymous data from already-deployed Wi-Fi infrastructure. We present a method applying geostatistical techniques to the access data acquired from Access Points (AP) in an open Wi-Fi network. Specifically, only the time series of the number of accesses per AP is required. Geostatistical methods produce a 3D high-quality spatial distribution representation of the people inside the building based on the interaction of their mobile devices with the APs. We report encouraging results obtained from data acquired at a building of Wroclaw University of Science and Technology.