Accurate and fast detection of typical fittings is the prerequisite of condition monitoring and fault diagnosis. At present, most successful fitting detectors are anchor-based, which are challenging to meet the requirements of edge deployment. In this paper, we propose a novel anchor-free method called HRM-CenterNet. Firstly, the lightweight MobileNetV3 is introduced into CenterNet to extract multi-scale features of different layers. In addition, the lightweight receptive field enhancement module is proposed for the deep layer features to further enhance the characterization power of global features and generate more accurate heatmaps. Finally, the high-resolution feature fusion network with iterative aggregation is designed to reduce the loss of spatial semantic information in subsampling and further improve the accuracy of small and occlusion objects. Experiments are carried out on the TFITS and PASCAL VOC datasets. The results show that the size of the network is more than 60% lower than that of CenterNet. Compared with other detectors, our method achieves comparable accuracy with all accurate models at a much faster speed and meets the performance requirements of real-time detection.
The accurate detection of computer room personnel can bring great convenience to computer room management and computer room inspection. Swin Transformer is used in object detection and achieves excellent detection performance. In this paper, Swin Transformer is used as the baseline to achieve accurate detection of computer room personnel. This paper mainly makes the following two contributions:1) In this paper, a practical self-attention method is designed. The channel interaction module is used in the self-attention calculation to solve the problem of local window self-attention lacking orientation awareness and location information. Reduce the size of input tokens through depth-wise convolution to reduce the complexity of self-attention calculation. 2) Use a balanced L1 loss and configure the weights of different stages of loss in the total loss function to solve the problem of imbalance between simple samples and difficult samples. Compared with the original Swin Transformer, the improved method improves the detection accuracy of mAP@0.5 by 3.2%.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.