基于混合特征与最小二乘支持向量机的山核桃鲜果识别  

Segmentation method for fresh hickory nut recognition based on mixed features and LS-SVM classifier

作　　者：章云[1,2] 周竹[1,2] 周素茵[1,2] ZHANG Yun ZHOU Zhu ZHOU Su-yin(School of Information Engineering, Zhejiang A＆F University, Lin＇an 311300, Zhejiang, China Zhejiang Provincial Key Laboratory of Forestry Intelligent Monitoring and Information Technology, Lin＇an 311300, Zhejiang, China)

机构地区：[1]浙江农林大学信息工程学院,浙江临安311300 [2]浙江省林业智能监测与信息技术研究重点实验室,浙江临安311300

出　　处：《中南林业科技大学学报》2016年第11期137-143,共7页Journal of Central South University of Forestry & Technology

基　　金：浙江省教育厅项目"基于机器视觉的山核桃智能识别研究"(Y201328071);浙江省自然科学基金项目"基于高光谱成像技术的锯材表面缺陷检测方法研究"(LQ13F050006)

摘　　要：针对颜色和背景相似的山核桃鲜果不易分离的问题,研究了自然环境下成熟山核桃鲜果的机器视觉识别方法。该算法以2×2像素的正方形区域作为分割单位,选择颜色比值B/R、R/G作为颜色特征,选择了基于灰度直方图的特征描述参数如均值、标准方差、平滑度、三阶矩、一致性和熵,以及基于灰度共生矩阵的特征描述参数角二阶矩、对比度和熵作为纹理特征,共同构成特征向量,采用LS-SVM算法建立了识别模型,并利用该模型对80幅顺光图像和50幅逆光图像进行测试。试验结果表明:该方法在顺光、逆光下的山核桃鲜果有效识别率分别为92.48%、88.15%,可为山核桃采摘机器人的研发提供技术参数。A fresh hickory nut （Carya cathayensis） recognition method based on mixed features composed of color features and texture features and least squares - support vector machine （LS-SVM） classifier is proposed for the fresh hickory nut that have similar color with leaves. The image is divided into 2 × 2 pixel blocks and the blocks is taken as the segmentation unit by the algorithm. Color feature （B/ R ratio, tUG ratio） is calculated based on RGB color model, and the texture features such as mean, standard variance, smoothness, three order moments, consistency and entropy based on gray histogram, and angular second moment, contrast ratio, and entropy based on gray level co-occurrence matrix （GLCM） are calculated in color channels of R, G, B, H and V. These sixty-two parameters are used as the input to the LS-SVM classifier. The model is applied to test 130 samples images taken in different illumination conditions. The results show that the recognition rate reaches 92.48% in front light conditions and 88.15% in back light conditions, respectively, It is feasible to use the algorithm in practical recognition of fresh hickory nut.

关 键 词：机器视觉 山核桃鲜果 图像分割 纹理特征 支持向量机 

分 类 号：S776.02[农业科学—森林工程]