基于深度学习的雾天非法采砂船只辨识方法  

Identification method of illegal sand mining vessels in foggy conditions based on deep learning

作　　者：张慧 王新猛 程占军 杨一涛 Zhang Hui;Wang Xinmeng;Cheng Zhanjun;Yang Yitao(School of Information Technology,Nanjing Police University,Nanjing 210023,China;Police Dog Technology Institute,Nanjing Police University,Nanjing 210012,China)

机构地区：[1]南京警察学院信息技术学院,南京210023 [2]南京警察学院警犬研究所,南京210012

出　　处：《电子测量与仪器学报》2025年第2期160-168,共9页Journal of Electronic Measurement and Instrumentation

基　　金：中央高校基本科研业务费专项资金(LGZD202407)项目资助。

摘　　要：为解决长江流域非法采砂船只试别困难、监测效率低、精度差等问题,提出一种基于深度学习的雾天非法采砂船只辨识方法。首先,提出改进的生成对抗网络对采集图像进行去雾处理,得到清晰水域图像。其中,利用融合特征注意力机制的生成器提取雾天环境下船只的复杂纹理特征;在判别器中加入谱归一化操作,解决网络训练梯度消失的问题;改进损失函数,引入循环一致性损失保证生成图像与原始图像间的结构一致性。其次,提出融合注意力机制的YOLOv8算法,有效增强网络对图像重要特征的提取能力,实现对清晰水域图像中非法采砂船只的精准定位与辨识。实验结果表明,本研究提出的改进生成对抗网络对图像的去雾效果较好,峰值信噪比(PSNR)与结构相似性(SSIM)分别为31.86和0.64,较Cycle GAN和GC-GAN算法分别提升了3.6%~13.1%、4.9%~56.1%。去雾后的图像经过融合注意力机制的YOLOv8算法处理,可实现对非法采砂船只的准确识别与准确定位,其mAP@0.5:0.95和帧率可达到89.6%、36 FPS,满足公安实战对精度与速度的要求,可有效提高长江流域非法采砂信息化、智能化监管与执法水平。To address the challenges of low monitoring efficiency and poor accuracy in identifying illegal sand mining vessels in foggy conditions,this study proposes a deep learning-based identification method.First,an improved generative adversarial network is employed to dehaze collected images,producing clear water area images.The generator integrates the feature attention mechanism to extract complex texture features of vessels in foggy environments,while spectral normalization is added to the discriminator to prevent gradient vanishing during training.Cycle consistency loss is introduced to ensure structural consistency between generated and original images.The CBAM attention mechanism is integrated into YOLOv8 algorithm to improve feature extraction,enabling precise localization and identification of illegal sand mining vessels in dehazed images.The improved GAN achieves superior dehazing performance,with PSNR and SSIM values of 31.86 and 0.64,representing 3.6%~13.1%and 4.9%~56.1%improvements over Cycle GAN and GCGAN,respectively.The YOLOv8 enhanced by CBAM achieves mAP@0.5:0.95 of 89.6%and FPS of 36 on dehazed images,meeting the accuracy and speed requirements for practical law enforcement.The proposed method effectively enhances the informatization and intelligence of illegal sand mining supervision and enforcement in the Yangtze River Basin.

关 键 词：非法采砂 图像去雾 生成对抗网络 目标检测 YOLOv8算法。 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] TN29[自动化与计算机技术—计算机科学与技术]