BP神经网络分类器在储粮害虫图像检索中的应用研究

针对储粮害虫图像检索特点,将BP神经网络分类器技术应用于储粮害虫图像检测和识别,成为近年来的一个研究热点.对采集到的包含储粮害虫的图像使用基于BP神经网络分类器的图像的储粮害虫图像检索方法,可以在线将储粮害虫从图像背景中分离出来,很好地保留了储粮害虫各个部分的纹理特征,并将信息分类存储以备查询.试验和仿真结果证明该算法不仅在线计算量小、便于实时控制,而且能够快速、高效地检测到储粮害虫的有效特征,同时进行分类检索.

Introducing Stored- Grain Pest Image Retrieval Based on BP Neural Network

According to the characters of stored - grain pest identification, a method of stored - grain pest image retrieval based on BP neural network has become a hot point of research nowadays. This method extracts visual fea-tures such as color and texture firstly, and then the proposed BP neural network identifies the image including pest in-formation. This method needs less computational effort and is easier applied.