基于卷积神经网络的辣椒识别

摘要

农产品检测技术一直以来都是农业领域研究的热点问题,但以往的识别的错误率都居高不下,该文采用了基于有深度学习机制的卷积神经网络方法来提高识别率.首先对采集到的图像进行预处理得到规范化的二值化图像,再利用Matlab软件进行神经网络的建模,利用其网络自学习能力进行训练与测试,通过仿真验证卷积神经网络对辣椒图像的精确识别率.并与传统BP神经网络进行比较,表明其具有很好的鲁棒性和泛化能力.%The agricultural Produce detection technology has always been a hot issue in the research field of agriculture,but previous recognition error rates are high,the paper uses a convolution neural network based depth learning mechanism approach to improve the recognition rate.Firstly,the collected images preprocessing to get normalized binary image,then using the Matlab software to establish neural network modeling,and using its self-learning ability for training and test and to verify accurate recognition rate of chili image through the simulation.It is compared with the traditional BP neural network which shows that it has good robustness and generalization ability.