摘要
利用机器视觉技术检测池塘水下自由活蟹的形态位置和数量分布信息,是实现自动投饵船精准变量投喂的关键。本文设计了一种基于联动扩展卷积神经网络的实时轻量型水下活蟹检测器。首先,针对水下图像模糊和色彩不均的特点,以及稀疏分解后不同频率图像的信息组成特点,分别进行K-SVD降噪和Retinex色彩校正;然后,采用联动扩展网络宽度、深度和分辨率方式来协调精度和效率的轻量级EfficientNet作为主干网络;引入复合缩放因子,对堆叠两层加权双向特征金字塔结构的高效融合特征网络和堆叠三层卷积模块的类别/边界框预测网络进行全局联动扩展,以构建适用于有限资源的小型活蟹检测器;最后,在类别/边界框预测网络中利用正交Softmax层替代完全连接的分类层,确保检测器可从小样本数据中学习更多的区分特征,有效缓解小样本检测的过拟合问题。采用自建的20625幅数据样本对检测模型进行训练和测试,实验表明,降噪、校正后的图像颜色均衡,且清晰度高,检测的平均交并比Iou提高近8个百分点。检测模型EfficientNet-Det0存储内存仅需15 MB,便可实现查准率96.21%和查全率94.86%,单幅图像检测延迟分别为10.6 ms(GPU)和35.0 ms(CPU)。浮点运算次数FLOPs减少至YOLOv3算法的1/15,CPU运行速度是其3倍,而准确性与YOLOv3算法相当,甚至略优。EfficientNet-Det0搭载在资源受限的自动投饵船上能够快速精准检测水下河蟹,并能实现对池塘自由活蟹分布的统计,为建立科学的投喂机制提供可靠的决策信息。
Using machine vision technology to detect the morphological position and quantity distribution information of underwater freely live crabs in the pond is a key step to realize the precise variable feeding for automatic feeding boats.However,it is very challenging to detect crabs quickly and reliably in the underwater images because of the large differences in posture,similarity joints of crabs,and the uneven color characteristics of the underwater images.For this reason,a real-time lightweight underwater live crab detector based on compound scaling convolutional neural networks was proposed.Firstly,aiming to the characteristics of underwater image blur and color imbalance,K-SVD denoising and Retinex color-correction enhancement were performed respectively according to the information composition characteristics with different frequencies after sparse decomposition.Secondly,a lightweight EfficientNet that perfectly coordinated the accuracy and efficiency by compound scaling the network width,depth and resolution was adopted as the backbone network.After that,a compound scaling factor was introduced to perform global overall compound scaling of the efficiently integrated feature network,which stacked two-layer weighted bi-directional feature pyramid structures,and the class/boundary-box prediction network that stacked three-layer convolution modules,to build a small-sized efficient detector for limited resources.In the class/boundary-box prediction network,an orthogonal Softmax layer was also adopted to replace the fully-connected classification layer to ensure that the detector can learn more distinguishing features from the small-sample dataset,which effectively alleviated the over-fitting problem of small-sample detection and improved the generalization ability of detectors.The model was trained and tested with self-built 20625 data samples.Experiments showed that images after denoising and correction were color-balanced and high-definition,and the detection Iou was increased by nearly 8 percentage points.The training detection model EfficientNet-Det0 can achieve 96.21%precision and 94.86%recall rate where only 15 MB storage memory was needed;the detection delay of a single image was only 10.6 ms(GPU)and 35.0 ms(CPU).Compared with the YOLOv3 algorithm,FLOPs was reduced to 1/15,and operating speed of CPU was increased by 2 times,yet the accuracy was comparable to or even better than YOLOv3.The EfficientNet-Det0 proposed was suitable for applying on resource-restricted automatic feeding bait to quickly and accurately detect underwater live crabs,which could realize the statistics of the distribution for freely live crabs in ponds,and provide reliable decision information for establishing scientific feeding mechanism.
作者
赵德安
曹硕
孙月平
戚浩
阮承治
ZHAO Dean;CAO Shuo;SUN Yueping;QI Hao;RUAN Chengzhi(School of Electrical and Information Engineering,Jiangsu University,Zhenjiang 212013,China;School of Mechanical and Electrical Engineering,Wuyi University,Wuyishan 354300,China)
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2020年第9期163-174,共12页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家自然科学基金项目(61903288)
江苏省自然科学基金项目(BK20170536)
福建省自然科学基金项目(2018J01471)
常州市现代农业科技项目(CE20192006)
江苏省高校优势学科建设项目(PAPD)。
关键词
自由活蟹检测
水下图像
颜色校正
卷积神经网络
联动扩展
freely live crab detection
underwater image
color correction
convolutional neural network
compound scaling
