基于图像处理技术与Android手机的小麦病害诊断系统
发布时间：2018-11-24 21:06
【摘要】：为了智能化农业的发展,提高农作物病害诊断水平,及时采取防治措施,提出一种基于图像处理技术与Android手机的病害智能诊断系统的设计方法.系统不受时间和地域限制,用户可以在具备网络覆盖的地方将采集到的病害图片发送至服务器端.服务器端接收到病害图片后,在HSI(horizintal situtation indicator)颜色空间对病害图像分割,利用颜色矩和灰度共生矩阵来提取病害颜色和纹理特征参数,并将优选的特征参数输入支持向量机识别,识别结果反馈给客户端.实验结果表明:该系统可以准确高效地识别出病害种类,有较强的实用性和推广应用前景.
[Abstract]:In order to improve the level of crop disease diagnosis and improve the development of intelligent agriculture, this paper puts forward a design method of intelligent disease diagnosis system based on image processing technology and Android mobile phone. The system is not limited by time and region, users can send the collected images to the server in the place with network coverage. After the server receives the disease image, the disease image is segmented in the HSI (horizintal situtation indicator) color space, and the color moment and gray level co-occurrence matrix are used to extract the disease color and texture feature parameters. The selected feature parameters are input into support vector machine (SVM) recognition, and the recognition results are fed back to the client. The experimental results show that the system can identify the diseases accurately and efficiently, and has strong practicability and application prospect.
【作者单位】： 郑州轻工业学院计算机与通信工程学院;
【基金】：国家自然科学基金资助项目(61302118) 河南省高校青年骨干教师资助计划(2010GGJS-114)
【分类号】：TP391.41
，
本文编号：2355056
[Abstract]:In order to improve the level of crop disease diagnosis and improve the development of intelligent agriculture, this paper puts forward a design method of intelligent disease diagnosis system based on image processing technology and Android mobile phone. The system is not limited by time and region, users can send the collected images to the server in the place with network coverage. After the server receives the disease image, the disease image is segmented in the HSI (horizintal situtation indicator) color space, and the color moment and gray level co-occurrence matrix are used to extract the disease color and texture feature parameters. The selected feature parameters are input into support vector machine (SVM) recognition, and the recognition results are fed back to the client. The experimental results show that the system can identify the diseases accurately and efficiently, and has strong practicability and application prospect.
【作者单位】： 郑州轻工业学院计算机与通信工程学院;
【基金】：国家自然科学基金资助项目(61302118) 河南省高校青年骨干教师资助计划(2010GGJS-114)
【分类号】：TP391.41
，
本文编号：2355056
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2355056.html
