Abstract
It is difficult for the autonomous underwater vehicle (AUV) to recognize targets similar to the environment in lacking data labels. Moreover, the complex underwater environment and the refraction of light cause the AUV to be unable to extract the complete significant features of the target. In response to the above problems, this paper proposes an underwater distortion target recognition network (UDTRNet) that can enhance image features. Firstly, this paper extracts the significant features of the image by minimizing the info noise contrastive estimation (InfoNCE) loss. Secondly, this paper constructs the dynamic correlation matrix to capture the spatial semantic relationship of the target and uses the matrix to extract spatial semantic features. Finally, this paper fuses the significant features and spatial semantic features of the target and trains the target recognition model through cross-entropy loss. The experimental results show that the mean average precision (mAP) of the algorithm in this paper increases by 1.52% in recognizing underwater blurred images.
Highlights
Underwater target recognition has difficulties in sample data collection and labeling, making it difficult to obtain labeled sample datasets
To address the above problems, this paper proposes an underwater distortion target recognition network (UDTRNet) via enhanced image features. e method allows fast recognition of underwater distortion targets in the absence of significant features
The three datasets, Cognitive Autonomous Diving Buddy (CADDY) underwater dataset, Underwater Image Enhancement Benchmark (UIEB), and Underwater Target dataset (UTD) are used for training and testing. e visual salient feature extraction model is trained by 13,000 unlabeled images
Summary
Underwater target recognition has difficulties in sample data collection and labeling, making it difficult to obtain labeled sample datasets. Unsupervised representational learning can extract significant features of images from unlabeled datasets and use them for target classification and detection tasks. Is method can improve the accuracy of underwater target recognition effectively in the case of insufficient tags. Unsupervised representational learning ignores some details of the image and only learns distinguishable features, which can improve the recognition speed of the algorithm. Currents, and complex underwater terrain can obscure the target. In this case, unsupervised representational learning is unable to extract the complete significant features for target recognition. Semantic relationship graphs can compensate for incomplete significant features
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have