Research on Chili Pepper Recognition and Localization Method Based on Binocular Vision
Main Article Content
Abstract
In this paper, the collected images are preprocessed by grayscale, image segmentation, binarization, etc., and then the parameters of the camera and the four types of coordinate transformation are analyzed by calibrating the camera, and then the AlexNet network and transfer learning pair are used The pepper was identified, and the correct rate of 90% was obtained, and the loss function was trained by cross-entropy with a loss value of 0.4, and finally the stereo matching and three-dimensional reconstruction were briefly introduced.
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
References
PENG Siyun, LUO Yi, XIE Ting, et al. Analysis and suggestions of "diamond model" of pepper industry competitiveness in Zunyi City[J]. Chili Magazine, 2019, 17(1):31-36
MAO Dong, JIANG Hua, et al. Development and current situation of Chaotian pepper industry in Zunyi[J]. China Vegetables, 2021, 2021(2):7-9
Ma Mengjie. Research on dynamic material identification method based on binocular stereo vision[D]. Chengdu: Southwest Jiaotong University, 2018
ZHU Lei. Research and implementation of vehicle detection, tracking and ranging method based on binocular vision[D]. Changsha: Hunan University, 2020
DING Wenkuan Chili Detection and Recognition Based on Convolutional Neural Network and Machine Vision [D]. Tianjin: Tianjin University of Technology, 2017
Leonard Brosgole. Dynamic size perception as a function of target location in egocentric space [J]. Bulletin of the Psychonomic Society， 1993， 31 (4):
Zhang Z. A Flexible New Technique for Camera Calibration [J]. Tpami ， 2000， 22 (11): 1330-1334.
WU Nianxiang, ZOU Huadong. Research on robot pose calibration technology based on stereo vision[J]. Journal of Jinggangshan University (Natural Science Edition), 2015,03:71-75.
DAI Yanfang,SONG Zhigang An Improved Image Segmentation Algorithm Implemented by MatLab[J]. Microprocessor, 2014, 05: 31-33