基于改进残差神经网络的家蚕日龄识别模型

doi:10.13733/j.jcam.issn.2095-5553.2024.02.037

摘要/Abstract

摘要： 家蚕日龄的准确识别有助于精准饲喂和动物福利,因此为准确识别家蚕生长时期中3龄第1天至5龄第7天,共14个日龄,在实际环境下采集特定家蚕品种,构建以14个日龄为单位的数据集。提出一种基于改进残差神经网络的Moga-ResNet,该方法在经典残差神经网络ResNet50的基础上,引入多阶门控机制以获取日龄图像的显著性特征。通过在同一个家蚕日龄数据集上开展模型训练与测试得到,Moga-ResNet的识别准确率为96.57%,F1值为96.57%,召回率为96.62%,与Swin Transformer、MobileNet v3、CSPNet和DenseNet四个经典模型的评价指标相比,Moga-ResNet在家蚕的日龄识别中具有较强的识别能力,可以为开展家蚕精准饲喂和数字化管理相关工作提供基础。

关键词: 家蚕, 日龄识别, 多阶门控机制, 残差神经网络

Abstract: Accurate identification of silkworm day-age contributes to precise feeding and animal welfare. In order to accurately identify 14 day-age from the first day of age 3 to the seventh day of age 5 in the growth period of silkworm, this paper constructs a dataset with 14 day-age units by collecting specific silkworm species in a real environment. This paper proposes Moga-ResNet method based on an improved residual neural network. This method introduces a multi-order gating mechanism to obtain the saliency features of day-age images based on the classical residual neural network ResNet50. Through model training and testing on the same domestic silkworm day-age dataset, the recognition accuracy of Moga-ResNet is 96.57%, the F1 value is 96.57%, and the recall rate is 96.62%. Compared with the evaluation indexes of four classic models, including Swin Transformer, MobileNet v3, CSPNet and DenseNet, Moga-ResNe achieved a stronger recognition ability in day-age recognition of domestic silkworm, which can provide a foundation for carrying out work related to precise feeding and digital management.

Key words: silkworm, day-age identification, multi-order gating mechanism, residual neural network

中图分类号:

TP391.4: S883.6

田丁伊, , 石洪康, , 祝诗平, , 陈肖, 张剑飞. 基于改进残差神经网络的家蚕日龄识别模型[J]. 中国农机化学报, 2024, 45(2): 259-266.

Tian Dingyi, , Shi Hongkang, , Zhu Shiping, , Chen Xiao, Zhang Jianfei. Day-age recognition model for the domestic silkworm based on improved residual neural network[J]. Journal of Chinese Agricultural Mechanization, 2024, 45(2): 259-266.

参考文献

［1］　朱慧敏. 硒添食对家蚕生长发育及硒蛋白相关基因表达的影响［D］. 镇江: 江苏科技大学, 2022.
Zhu Huimin. Effects of selenium supplementation on growth and development of silkworm and expression of selenoprotein-related genes ［D］. Zhenjiang: Jiangsu University of Science and Technology, 2022.
［2］　张雨丽, 李豪, 樊银凤, 等. 全龄人工饲料育与桑叶育家蚕生丝的蛋白质成分比较［J］. 蚕业科学, 2022, 48(3): 228-236.
Zhang Yuli, Li Hao, Fan Yinfeng, et al. Comparative analysis of protein composition of raw silk from silkworm reared on artificial diet and mulberry leaf in all instars ［J］. Acta Sericologica Sinica, 2022, 48(3): 228-236.
［3］　蔡幼民. 家蚕人工饲料育研究工作回顾［J］. 蚕业科学, 2022, 48(1): 1-6.
Cai Youmin. A review of our research on artificial diet rearing of silkworm ［J］. Acta Sericologica Sinica, 2022, 48(1): 1-6.
［4］　逄玉军, 马彦辉, 程国辉, 等. 辽宁省柞蚕留根单母的时代演进和生产要点［J］. 蚕桑茶叶通讯, 2020(1): 7-8.
［5］　孙亮, 柯宇航, 刘辉, 等.计算机视觉技术在植物病害识别上的研究进展［J］. 热带生物学报, 2022, 13(6): 651-658.
Sun Liang, Ke Yuhang, Liu Hui, et al. Advances in recognition of plant diseases based on computer vision ［J］. Journal of Tropical Biology, 2022, 13(6): 651-658.
［6］　任治洲, 谢云, 晁志恒, 等. 基于物联网与深度学习技术的农作物生长状况远程动态监测系统［J］. 物联网技术, 2022, 12(8): 15-18, 21.
［7］　杨涛, 李晓晓. 机器视觉技术在现代农业生产中的研究进展［J］. 中国农机化学报, 2021, 42(3): 171-181.
Yang Tao, Li Xiaoxiao. Research progress of machine vision technology in modern agricultural production ［J］. Journal of Chinese Agricultural Mechanization, 2021, 42(3): 171-181.
［8］　何雨霜, 王琢, 王湘平, 等. 深度学习在农作物病害图像识别中的研究进展［J］. 中国农机化学报, 2023, 44(2): 148-155.
He Yushuang, Wang Zhuo, Wang Xiangping, et al. Research progress of deep learning in crop disease image recognition ［J］. Journal of Chinese Agricultural Mechanization, 2023, 44(2): 148-155.
［9］　石洪康, 田涯涯, 杨创, 等. 基于卷积神经网络的家蚕幼虫品种智能识别研究［J］. 西南大学学报(自然科学版), 2020, 42(12): 34-45.
Shi Hongkang, Tian Yaya, Yang Chuang, et al. Research on intelligent recognition of silkworm larvae races based on convolutional neural networks ［J］. Journal of Southwest University (Natural Science Edition), 2020, 42(12): 34-45.
［10］　石洪康, 肖文福, 黄亮, 等. 基于卷积神经网络的家蚕病害识别研究［J］. 中国农机化学报, 2022, 43(1): 150-157.
Shi Hongkang, Xiao Wenfu, Huang Liang, et al. Research on recognition of silkworm diseases based on convolutional neural network ［J］. Journal of Chinese Agricultural Mechanization, 2022, 43(1): 150-157.
［11］　于业达, 高鹏飞, 赵一舟, 等. 基于深度卷积神经网络的蚕蛹雌雄自动识别［J］. 蚕业科学, 2020, 46(2): 197-203.
Yu Yeda, Gao Pengfei, Zhao Yizhou, et al. Automatic identification of female and male silkworm pupa based on deep convolution neural network ［J］. Acta Sericologica Sinica, 2020, 46(2): 197-203.
［12］　陶丹. 基于机器视觉的家蚕蛹雌雄识别研究［D］. 重庆: 西南大学, 2019.
Tao Dan. Sex discrimination study of silkworm pupae based on machine vision technology ［D］. Chongqing: Southwest University, 2019.
［13］　何锐敏, 郑可锋, 尉钦洋, 等. 基于改进Mask R-CNN模型的工厂化养蚕蚕体识别与计数［J］. 智慧农业(中英文), 2022, 4(2): 163-173.
He Ruimin, Zheng Kefeng, Wei Qinyang, et al. Identification and counting of silkworms in factory farm using improved mask R-CNN model ［J］. Smart Agriculture, 2022, 4(2): 163-173.
［14］　Wen C, Wen J, Li J, et al. Lightweight silkworm recognition based on multi-scale feature fusion ［J］. Computers and Electronics in Agriculture, 2022, 200: 107234.
［15］　李时杰, 孙卫红, 梁曼, 等. 基于深度学习的蚕茧种类实时检测系统设计［J］. 上海纺织科技, 2021, 49(11): 53-55, 58.
Li Shijie, Sun Weihong, Liang Man, et al. Design of real-time detection system for silkworm cocoon species based on deep learning ［J］. Shanghai Textile Science & Technology, 2021, 49(11): 53-55, 58.
［16］　张友洪, 沈以红, 肖文福, 等. 家蚕杂交组合芳·绣 × 白·春的选配［J］. 蚕业科学, 2014, 40(6): 1017-1023.
Zhang Youhong, Shen Yihong, Xiao Wenfu, et al. Selective breeding of Bombyx mori cross combination Fang· Xiu × Bai· Chun ［J］. Acta Sericologica Sinica, 2014, 40(6): 1017-1023.
［17］　He K, Zhang X, Ren S, et al. Deep residual learning for image recognition ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 770-778.
［18］　Li S, Wang Z, Liu Z, et al. Efficient multi-order gated aggregation network ［J］. arXiv preprint arXiv: 2211.03295, 2022.
［19］　Liu Z, Lin Y, Cao Y, et al. Swin transformer: Hierarchical vision transformer using shifted windows ［C］. Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 10012-10022.
［20］　Howard A G, Zhu M, Chen B, et al. MobileNets: Efficient convolutional neural networks for mobile vision applications ［J］. arXiv preprint arXiv: 1704.04861, 2017.
［21］　Wang C Y, Liao H Y M, Wu Y H, et al. CSPNet: A new backbone that can enhance learning capability of CNN ［C］. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2020: 390-391.
［22］　Huang G, Liu Z, Van DerMaaten L, et al. Densely connected convolutional networks ［C］. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 4700-4708.
（上接第226页）
Wu Xiru, Xue Qiwei. 3D vehicle detection for unmanned driving system based on lidar ［J］. Optics and Precision Engineering, 2022, 30(4): 489-497.
［13］　尚业华, 张光强, 孟志军, 等. 基于欧氏聚类的三维激光点云田间障碍物检测方法［J］. 农业机械学报, 2022, 53(1): 23-32.
Shang Yehua, Zhang Guangqiang, Meng Zhijun, et al. Field obstacle detection method of 3D Lidar point cloud based on euclidean clustering ［J］. Transactions of the Chinese Society for Agricultural Machinery, 2022, 53(1): 23-32.
［14］　Yan Z, Duckett T, Bellotto N. Online learning for 3D LiDAR-based human detection: Experimental analysis of point cloud clustering and classification methods ［J］. Autonomous Robots, 2020, 44: 147-164.
［15］　刘亮. 基于激光雷达的结构化道路环境感知及避障路径规划方法研究［D］. 西安: 长安大学, 2021.
Liu Liang. Research on environmental awareness and path planning of structured road based on Lidar ［D］. Xian: Changan University, 2021.
［16］　刘浩然, 范伟伟, 徐永胜, 等. 基于无人机激光雷达点云数据的单木分割研究［J］. 中南林业科技大学学报, 2022, 42(1): 45-53.
Liu Haoran, Fan Weiwei, Xu Yongsheng, et al. Research on single tree segmentation based on UAV LiDAR point cloud data ［J］. Journal of Central South University of Forestry & Technology, 2022, 42(1): 45-53.
［17］　赵星阳. 基于车载激光雷达的路面分割与障碍物检测关键技术研究［D］. 合肥: 合肥工业大学, 2021.
Zhao Xingyang. Research on road segmentation and obstacle detection of autonomous vehicle based on Lidar ［D］.Hefei: Hefei University of Technology, 2021.
［18］　Tian Y, Song W, Chen L, et al. A fast spatial clustering method for sparse LiDAR point clouds using GPU programming ［J］. Sensors, 2020, 20(8): 2309.
［19］　Mohd Romlay M R, Mohd Ibrahim A, Toha S F, et al. Novel CE-CBCE feature extraction method for object classification using a low-density LiDAR point cloud ［J］. PloS one, 2021, 16(8): e0256665.
［20］　Nguyen H T, Lee E H, Bae C H, et al. Multiple object detection based on clustering and deep learning methods ［J］. Sensors, 2020, 20(16): 4424.
［21］　Gamal A, Wibisono A, Wicaksono S B, et al. Automatic LiDAR building segmentation based on DGCNN and euclidean clustering ［J］. Journal of Big Data, 2020, 7(1): 1-18.
［22］　Gao F, Li C, Zhang B. A dynamic clustering algorithm for LiDAR obstacle detection of autonomous driving system ［J］. IEEE Sensors Journal, 2021, 21(22): 25922-25930.
［23］　张博闻. 面向开放场景的自动驾驶激光雷达目标检测技术研究［D］. 重庆: 重庆大学, 2020.
Zhang Bowen. Research on automatic lidar target detection technology for open scene ［D］. Chongqing: Chongqing University, 2020.