Computer Vision and Intelligent Sensing Based on Pattern Recognition

A novel approach to detect and decode direct-part-marked, low-contrast data matrix codes on polymer-based selective laser sintering manufactured parts, which is able to work on lightweight devices, is presented. Direct-part marking is a concept for labeling parts directly, which can be carried out during the additive manufacturing’s design process. Because of low contrast in polymer-based selective laser sintering manufactured parts, it is a challenging task to detect and read codes on unicolored parts. To achieve this, at first, codes are located using a deep-learning-based approach. Afterwards, the calculated regions of interest are passed into an image encoding network in order to compute readable standard data matrix codes. To enhance the training process, rendered images, improved with a generative adversarial network, are used. This process fulfills the traceability task in assembly line production and is suitable for running on mobile devices such as smartphones or cheap sensors placed in the assembly line. The results show that codes can be localized with 97.38% mean average precision, and a readability of 89.36% is achieved. Full article

The purpose of image dehazing is to remove the interference from weather factors in degraded images and enhance the clarity and color saturation of images to maximize the restoration of useful features. Single image dehazing is one of the most important tasks in the field of image restoration. In recent years, due to the progress of deep learning, single image dehazing has made great progress. With the success of Transformer in advanced computer vision tasks, some research studies also began to apply Transformer to image dehazing tasks and obtained surprising results. However, both the deconvolution-neural-network-based dehazing algorithm and Transformer based dehazing algorithm magnify their advantages and disadvantages separately. Therefore, this paper proposes a novel Transformer–Convolution fusion dehazing network (TCFDN), which uses Transformer’s global modeling ability and convolutional neural network’s local modeling ability to improve the dehazing ability. In the Transformer–Convolution fusion dehazing network, the classic self-encoder structure is used. This paper proposes a Transformer–Convolution hybrid layer, which uses an adaptive fusion strategy to make full use of the Swin-Transformer and convolutional neural network to extract and reconstruct image features. On the basis of previous research, this layer further improves the ability of the network to remove haze. A series of contrast experiments and ablation experiments not only proved that the Transformer–Convolution fusion dehazing network proposed in this paper exceeded the more advanced dehazing algorithm, but also provided solid and powerful evidence for the basic theory on which it depends. Full article

At many construction sites, whether to wear a helmet is directly related to the safety of the workers. Therefore, the detection of helmet use has become a crucial monitoring tool for construction safety. However, most of the current helmet wearing detection algorithms are only dedicated to distinguishing pedestrians who wear helmets from those who do not. In order to further enrich the detection in construction scenes, this paper builds a dataset with six cases: not wearing a helmet, wearing a helmet, just wearing a hat, having a helmet, but not wearing it, wearing a helmet correctly, and wearing a helmet without wearing the chin strap. On this basis, this paper proposes a practical algorithm for detecting helmet wearing states based on the improved YOLOv5s algorithm. Firstly, according to the characteristics of the label of the dataset constructed by us, the K-means method is used to redesign the size of the prior box and match it to the corresponding feature layer to increase the accuracy of the feature extraction of the model; secondly, an additional layer is added to the algorithm to improve the ability of the model to recognize small targets; finally, the attention mechanism is introduced in the algorithm, and the CIOU_Loss function in the YOLOv5 method is replaced by the EIOU_Loss function. The experimental results indicate that the improved algorithm is more accurate than the original YOLOv5s algorithm. In addition, the finer classification also significantly enhances the detection performance of the model. Full article

Contrastive learning has received increasing attention in the field of skeleton-based action representations in recent years. Most contrastive learning methods use simple augmentation strategies to construct pairs of positive samples. When using such pairs of positive samples to learn action representations, deeper feature information cannot be learned, thus affecting the performance of downstream tasks. To solve the problem of insufficient learning ability, we propose an asymmetric data augmentation strategy and attempt to apply it to the training of 3D skeleton-based action representations. First, we carefully study the different characteristics presented by different skeleton views and choose a specific augmentation method for a certain view. Second, specific augmentation methods are incorporated into the left and right branches of the asymmetric data augmentation pipeline to increase the convergence difficulty of the contrastive learning task, thereby significantly improving the quality of the learned action representations. Finally, since many methods directly act on the joint view, the augmented samples are quite different from the original samples. We use random probability activation to transform the joint view to avoid extreme augmentation of the joint view. Extensive experiments on NTU RGB + D datasets show that our method is effective. Full article

Inventory is the basis of business activities; inventory management helps industries keep their inventories stocked with reasonable quantities, which ensures consumers demand while minimizing storage costs. The traditional manual inventory management has low efficiency and a high labor cost. In this paper, we used improved YOLOv3 to detect the cups stored on the warehouse shelves and counted their numbers to realize automated inventory management. The warehouse images are collected by the camera and transmitted to the industrial computer, which runs the YOLOv3 network. There are three feature maps in YOLOv3, the two smaller feature maps and the structure behind them are removed, and the k-means algorithm is used to optimize the default anchor size. Moreover, the detection range is limited to a specified area. Experiments show that, by eliminating those two feature maps, the network parameter is reduced from 235 MB to 212 MB, and detection FPS is improved from 48.15 to 54.88 while mAP is improved from 95.65% to 96.65% on our test dataset. The new anchors obtained by the k-means algorithm further improve the mAP to 96.82%. With those improvements, the average error rate of detection is reduced to 1.61%. Restricted detection areas eliminate irrelevant items to ensure the high accuracy of the detection result. The accurately counted number of cups and its change provide significant data for inventory management. Full article