基于数据增强的高原鼠兔目标检测  

Target detection of Ochotona curzoniae based on the data augmentation

作　　者：陈海燕[1] 甄霞军 赵涛涛 CHEN Haiyan;ZHEN Xiajun;ZHAO Taotao(School of Computer and Communication,Lanzhou University of Technology,Lanzhou Gansu 730050,China)

机构地区：[1]兰州理工大学计算机与通信学院,甘肃兰州730050

出　　处：《湖南农业大学学报（自然科学版）》2022年第4期496-500,共5页Journal of Hunan Agricultural University(Natural Sciences)

基　　金：国家自然科学基金项目(62161019、62061024)。

摘　　要：针对基于卷积神经网络的高原鼠兔目标检测模型在实际应用中缺乏训练数据的问题,提出一种前景与背景融合的数据增强方法:首先对训练集数据进行前景和背景的分离,对分离的前景作图像随机变换,对分离的背景用背景像素随机覆盖,得到前景集合和背景集合;从前景集合和背景集合中随机选取前景和背景,进行像素加融合;再从训练集中随机选取样本,将标注边界框区域采用剪切粘贴方法融合到训练图像的随机位置,得到增强数据集。采用两阶段的弱监督迁移学习训练模型,第一阶段在增强数据集上对模型预训练;第二阶段在原始训练集上微调预训练模型,得到检测模型。对自然场景下高原鼠兔目标检测的结果表明:在相同的试验条件下,基于前景与背景融合数据增强的目标检测模型的平均精度优于未数据增强、Mosaic和CutOut数据增强的目标检测模型;基于前景、背景融合数据增强的目标检测模型的最优平均精度为78.4%,高于Mosaic的72.60%、Cutout的75.86%和Random Erasing的77.4%。Aiming at the problem that the Ochotona curzoniae target detection model based on convolutional neural network lacks training data in practical application,a data augmentation method is proposed by the fusion of foreground and background.Firstly,separate the foreground and the background of the training data,with image transforming the separated foreground randomly and covering the separated background by background pixels,to obtain the foreground set and the background set,respectively.The foreground and background are randomly selected from the foreground set and the background set,respectively and are fused based on pixel addition.Then randomly select a sample from the training set,and use the cut-and-paste method to fuse the labeled bounding box area of the selected sample to the training images'random positions to obtain an augmented data set.A two-stage weakly supervised transfer learning was used as the train the model.The first stage pre-trains the model dependent on the augmented data set.The second stage fine-tunes the pre-training model to obtain the detection model.Under the same experimental conditions,the experimental results of the target detection of Ochotona curzoniae in natural scenes show that the average accuracy of the target detection model based on this method is better than that of the target detection model without data augmentation,Mosaic,and Cutout data augmentation.The optimal AP of the target detection model based on data augmentation method by the fusion of foreground and background is 78.4%,which is higher than 72.6%of Mosaic method,75.86%of Cutout method,and 77.4%of Random Erasing method.

关 键 词：高原鼠兔 样本缺乏 数据增强 迁移学习 样本平衡 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]