Content of Target Detection in our journal

Published in last 1 year |  In last 2 years |  In last 3 years |  All Please wait a minute... For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails Select Complex Road Target Detection Algorithm Based on Improved YOLOv5 WANG Pengfei, HUANG Hanming, WANG Mengqi Computer Engineering and Applications    2022, 58 (17): 81-92.   DOI: 10.3778/j.issn.1002-8331.2205-0158 Abstract （376）      PDF（pc） （1217KB）（228）       Knowledge map    Save Aiming at the problem of false detection and missed detection caused by dense occluded targets and small targets in complex road background, a complex road target detection algorithm based on improved YOLOv5 is proposed. Firstly, Quality Focal Loss is introduced, which combines the classification score with the quality prediction of location to improve the positioning accuracy of dense occluded targets. Secondly, a shallow detection layer is added as the detection layer of smaller targets, the three-scale detection of the original algorithm is changed to four-scale detection, and the feature fusion part is also improved accordingly, which improves the learning ability of the algorithm to the features of small targets. Then, based on the feature fusion idea of weighted bidirectional feature pyramid network（BiFPN）, a de-weighted BiFPN is proposed, which makes full use of deep, shallow and original feature information, strengthens feature fusion, reduces the loss of feature information in the process of convolution, and improves the detection accuracy. Finally, the convolution block attention module（CBAM） is introduced to further improve the feature extraction ability of the algorithm and make the algorithm pay more attention to useful information. The experimental results show that the detection accuracy of the improved algorithm in this paper on the public autopilot data set KITTI and the self-made rider helmet data set Helmet reaches 94.9% and 96.8% respectively, which is 1.9 percentage points and 2.1 percentage points higher than the original algorithm, and the detection speed reaches 69 FPS and 68 FPS respectively. It has better detection accuracy and real-time performance. At the same time, compared with some mainstream target detection algorithms, the improved algorithm in this paper also has some advantages. Reference | Related Articles | Metrics Select Remote Sensing Image Target Detection Algorithm Based on Residual Shrinkage Network GAO Ye, GUO Songyi, SHE Xiangyang Computer Engineering and Applications    2022, 58 (17): 93-100.   DOI: 10.3778/j.issn.1002-8331.2109-0290 Abstract （144）      PDF（pc） （1360KB）（72）       Knowledge map    Save Aiming at the problems of complex background，more noise，dense arrangement of small targets and large difference of target scale in remote sensing images，a remote sensing image target detection algorithm based on improved channel attention and residual shrinkage network is proposed in this paper. The algorithm uses convolutional neural network，takes YOLOV3 model as the basic network，selects mosaic image enhancement for data preprocessing，reconstructs the feature extraction network by using the depth residual shrinkage module，constructs the spatial pyramid pooling fusion layer by combining the channel attention mechanism and combination pooling，calculates the positioning loss by using CIOU，and finally realizes the target detection of remote sensing image. Experimental results show that compared with the original algorithm，the overall map of the improved algorithm is improved from 89.2% to 92.2%，and better performance is obtained. Reference | Related Articles | Metrics Select Target Detection Algorithm Based on Multi-Scale Combined Weight Distribution CUI Jingwen, MA Jie, ZHANG Yu Computer Engineering and Applications    2022, 58 (17): 101-110.   DOI: 10.3778/j.issn.1002-8331.2109-0438 Abstract （120）      PDF（pc） （1225KB）（44）       Knowledge map    Save For the problem of SSD（single shot multibox detector） identification in complex traffic environments, this paper proposes a target detection improvement algorithm based on multi-scale feature complementarity and combined weight distribution（multi-scale feature complementary fusion and key feature information mining SSD, MK-SSD）. The proposed algorithm first designs multi-scale feature complementary modules using cross stage partial network and builds multi-path feature fusion networks to effectively improve the feature extraction capability of shallow networks for small targets. Secondly, the combined weight distribution module is designed to combine the perceptual domain with the key information mining to more efficiently use the key feature information and suppress the attention of the non-key information. Finally, the prediction network is improved using lightweight residual blocks to improve the target detection capability. After experimental analysis, the average accuracy of the improved algorithm reaches 89.64% on the homemade traffic sign dataset. While ensuring real-time, it has higher detection accuracy compared with YOLO series and SSD series algorithms, and can detect small targets of most SSD network leakage detection. Reference | Related Articles | Metrics Select Remote Sensing Image Small Target Detection Method Using Simulation Image as Template CAO Yaming, XIAO Qi, YANG Zhen Computer Engineering and Applications    2022, 58 (17): 111-119.   DOI: 10.3778/j.issn.1002-8331.2103-0568 Abstract （199）      PDF（pc） （2144KB）（39）       Knowledge map    Save With the continuous progress of sensor technology and aerial remote sensing technology, the quality and quantity of remote sensing images have been greatly improved. Target detection in remote sensing images is a basic problem in understanding and analyzing remote sensing images. Aiming at the problems that neural network is difficult to extract enough effective features in remote sensing image small target detection task, and remote sensing small target is easily blocked by cloud and fog, a method based on simulation image template matching is proposed, which is successfully applied to remote sensing image small target detection task by feature fusion. The simulation image generated by imaging simulation technology contains more features of remote sensing small target, such as geometry, material and so on. In the process of combining with deep learning, more features can improve the accuracy of neural network for small target detection in remote sensing image. The results show that the template matching method based on the simulation image is applied to the deep learning, and it achieves good detection results for small targets in remote sensing images, especially for small targets disturbed by the weather such as clouds and fog. Reference | Related Articles | Metrics Select Object Detection Based on Dual Attention Mechanism Combined with Discriminant Correlation Analysis ZHAO Shan, ZHENG Ailing Computer Engineering and Applications    2022, 58 (17): 120-129.   DOI: 10.3778/j.issn.1002-8331.2109-0452 Abstract （108）      PDF（pc） （2063KB）（64）       Knowledge map    Save For the problems of low target recognition rate and missing detection of some small targets in the model of two-stage target detection algorithm, a target detection algorithm with dual attention mechanism based on discriminant correlation analysis is proposed in this paper. In order to maximize correlation between the corresponding features in two feature sets and differences between different types of features, the Faster R-CNN backbone network is improved and discriminant correlation analysis technology is introduced. It can ensure the interaction of information and effectively alleviate the problem of insufficient feature extraction ability in the conventional feature fusion method. At the same time, a residual dual attention mechanism combined with the residual structure is constructed, which aims to extract the deep-level feature so as to compensate for the weakening of high-resolution information after deep CNN. Meantime, the mixed convolutional layer is adopted to expand receptive field while reducing information loss to maximize the feature extraction performance of network. PASCAL VOC2007, KITTI and Portrait are adopted to train the network, and the proposed algorithm model is compared with multiple classic target detection algorithms. Experimental results demonstrate that the proposed algorithm has high detection accuracy. Reference | Related Articles | Metrics