Research On Underwater Fish Image Enhancement And Detection Based On Deep Learning

Automatic detection and identification of fish targets from underwater videos and images is of great significance for the assessment of fishery resources and ecological environment monitoring.However,due to the poor quality of underwater images and unconstrained fish movement,traditional hand-designed feature extraction methods or the convolutional neural network（CNN）-based object detection algorithms cannot meet the detection requirements in real underwater scenes.In the process of achieving underwater fish target detection,it is first necessary to solve the problem of low-resolution,low-quality underwater fish images,and the problem of fish target detection after the reconstructed high-resolution fish images are obtained.This article focuses on the main technical difficulties of underwater fish target detection:（1）In view of the relatively low resolution and contrast of underwater images,this paper proposes an image SR reconstruction method based on a generative adversarial network with a residual dense architecture.First,before Re LU activation,the number of feature channels is expanded by a factor of 6～9 using a 1×1 convolutional layer,which improves the utilization of shallow information.Next,the original discriminator is replaced with a relativistic average discriminator,thereby improving the authenticity of the discriminative network.Finally,preactivation features are used to improve the perceptual loss,thus providing stronger monitoring for brightness consistency and texture restoration.Experimental results show that the proposed algorithm improves the utilization of shallow information in a deep network.Structural similarity（SSIM）index evaluations show that the overall utilization of shallow information is increased by 105.52%.In addition,the average runtime is 0.42 sec/frame,nearly3.6 times faster than those of traditional methods.Moreover,the recovered images have an average natural image quality evaluator value of 3.4 and high perceptual quality.（2）Aiming at the problem of fish detection in complex underwater environments,this paper proposes a composite backbone network Two-stage target detection network Composite Backbone and Enhanced Path Aggregation Network（CBEPANet）.By improving the residual network（Res Net）,a new composite backbone network（Cbresnet）is designed to learn the scene change information（source domain style）,which is caused by the differences in the image brightness,fish orientation,seabed structure,aquatic plant movement,fish species shape and texture differences.Thus,the interference of underwater environmental information on the object characteristics is reduced,and the output of the main network to the object information is strengthened.In addition,to better integrate the high and low feature information output from Cbresnet,the enhanced path aggregation network（EPANet）was also designed to solve the insufficient utilization of semantic information caused by linear upsampling.The experimental results show that the average precision（AP）_0.5:0.95,AP₅₀ and average recall（AR）_max=10 of the proposed Composited Fish Net is 75.2%,92.8%and 81.1%,respectively.The composite backbone network enhances the characteristic information output of the detected object and improves the utilization of characteristic information.