【摘要】 对于田间青椒的识别,利用图像处理,提出一种采用BP神经网络的识别方法。将青椒的形状系数、紧密度、长宽比、伸长度和主轴周长比等5个归一化变量作为神经网络的输入,通过调节神经网络的参数值对神经网络进行优化训练,利用训练后的网络就可以对处理后的青椒图像进行识别。实验结果表明,训练后的神经网络能够以较高精度进行田间青椒的识别。更多还原

【Abstract】 In this paper,a method that bases on the BP neural network is introduced in the process of Identification of Green Pepper Image in the Field. Through the image processing,the five shape characteristic as the BP neural network input,such as,the form factor,the tightness,the length and breadth rate,the elongation,the main axle perimeter rate is obtained. The green pepper image can be identified by the neural network that is trained through adjusting its parameter value. The result of the experiment indicates that the field green pepper fruit can be identified well by training the BP neural network.更多还原