夜间动物图像自监督学习增强与检测方法OA北大核心CSTPCD
Self-supervised learning enhancement and detection methods for nocturnal animal images
为了解决动物夜间实时监测所面临的图像曝光度低、对比度低、特征提取困难等问题,通过研究轻量化自监督深度神经网络Zero-Denoise和改进型YOLOv8模型,来进行夜间动物目标的图像增强与精准识别.首先,通过轻量化的PDCE-Net进行第一阶段快速增强.提出了一个新的光照损失函数,并利用参数可调的Gamma校正原图与快速增强图,在基于Retinex原理和最大熵理论的PRED-Net中进行第二阶段的重增强.然后,改进YOLOv8模型,并对重增强后的…查看全部>>
In order to solve the problems of low image exposure,low contrast and difficulty of feature extrac-tion in real-time animal monitoring at night,we proposed a lightweight self-supervised deep neural network Zero-Denoise and an improved YOLOv8 model for image enhancement and accurate recognition of nocturn-al animal targets.The first stage of rapid enhancement was performed by lightweight PDCE-Net.A new lighting loss function was proposed,and the second stage …查看全部>>
王驰;沈晨;黄庆;张国峰;卢汉;陈金波
上海大学机电工程与自动化学院,上海 200444上海大学机电工程与自动化学院,上海 200444中国航空工业集团公司洛阳电光设备研究所,河南洛阳 471023上海大学机电工程与自动化学院,上海 200444上海大学机电工程与自动化学院,上海 200444上海大学机电工程与自动化学院,上海 200444
计算机与自动化
夜间动物监测低光增强自监督学习Retinex低光去噪
nocturnal animal detectionlow-light enhancementself-supervised learningRetinexlow-light denoising
《中国光学（中英文）》 2024 (5)
1087-1097,11
国家自然科学基金项目(No.62175144)北京市航空智能遥感装备工程技术研究中心开放基金课题(No.AIRSE20233) Supported by the National Natural Science Foundation of China(No.62175144)Beijing Engineering Research Center of Aerial Intelligent Remote Sensing Equipments Open Fund(No.AIRSE20233)
评论