A Novel Attention-Based Lightweight Network for Multiscale Object Detection in Underwater Images

Abstract

Underwater images have low quality, and underwater targets have different sizes. The mainstream target detection networks cannot achieve good results in detecting objects from underwater images. In this study, a lightweight underwater multiscale target detection model with an attention mechanism is designed to solve the above problems. In this model, MobileNetv3 is used as the backbone network for preliminary feature extraction. The lightweight feature extraction module (LFEM) pays attention to the feature map at the channel and space levels. The features with large weights are promoted, while the features with small weights are suppressed. Meanwhile, cross-group information exchange enriches the semantic information and location information of the objects. The context aggregation module (CIAM) pools the extracted feature maps to obtain feature pyramids, and it uses the upsampling-feature refinement-cascade addition (URC) method to effectively fuse global context information and enhance the feature representation. The scale normalization for feature pyramids (SNFP) performs adaptive multiscale perception and multianchor detection on feature maps to cover objects of different sizes and realize multiscale object detection in underwater images. The proposed network can realize lightweight feature extraction, effectively handle the global relationship between the underwater scene and the object while expanding the receptive field, traverse the objects of different scales, and achieve adaptive multianchor detection of multiscale objects in underwater images. The experimental results indicate that our method achieves an average accuracy of 81.94% and a detection speed of 44.3 FPS on a composite dataset. Also, our method is better than the mainstream object detection networks in terms of detection accuracy, lightweight design, and real-time performance.

Underwater Small Target Detection Based on YOLOX Combined with MobileViT and Double Coordinate Attention

“…Yeh et al [28] proposed a lightweight deep neural network for joint learning of underwater object detection and color conversion. Wang et al [29] applied a novel attention-based lightweight network for multiscale object detection in underwater images. The downside is that the use of lightweight methods inevitably leads to reductions in detection accuracy.…”

Estimating catch rates in real time: Development of a deep learning based Nephrops (Nephrops norvegicus) counter for demersal trawl fisheries

“…For instance, MobileNets are efficient models developed to be used in hardware with limited computational resources (Howard et al, 2017) and can be used as a standalone classifier for animal classification in underwater images (Liu et al, 2019). Together with two other improved versions (Sandler et al, 2018;Howard et al, 2019) and single shot object detectors (SSD), they have more diverse applications such as detection of sea cucumbers (Yao et al, 2019), underwater objects with different scales (Zhang et al, 2021;Wang et al, 2022b), and Nephrops burrows (Naseer et al, 2020).…”

Receptive field enhancement and attention feature fusion network for underwater object detection

“…They perform excellently in land optical environments and achieve very good results on various land optical datasets. However, due to the complex and changeable underwater environment, 4 images taken by the underwater vision system are affected by high noise, low visibility, edge blur, low contrast, color bias, and shadowing from passive light sources 5 , 6 . These effects often lead to image blur and texture distortion, 7 , 8 which have a great impact on the accuracy of object detection.…”