阵列式传感器的触觉图像识别技术

通过对安装于机器人手爪上的半导体应变式阵列触觉传感器系统所采集到的触觉图像的分析，提出了基于模糊聚类原理的触觉图像统计特征识别技术。通过分析像素灰度矩阵信息，提取图像独立的统计特征量，建立相应的图像统计特征模型空间；由此，计算出各图像类的模糊相似系数矩阵，再运用聚类分析法将其改造成为模糊等价矩阵，划分出图像等价类，进而实现对目标图像的识别；实验结果表明，该方法能获得分布压、接触总力等信息，并能达到很好的触觉图像识别效果。

Image Recognition Based on Tactile Sensor Array

The tactile image collected by robotic paw tactile sensor array which use semiconductor strain-gauge is analyzed in details. Based on fuzzy cluster analysis, a scheme for recognizing tactile image is presented. To analyze the spatial knowledge of pixels' hue matrix, and extract static features of image, a fuzzy similitude matrix is defined to describe the image's feature space. Introducing the fuzzy cluster analysis, a corresponding fuzzy equivalence matrix is obtained over the fuzzy similitude matrix by computing the transferred package. Under the biggest threshold, images are classified to two groups. The experiment shows that the proposed approach is effective on image recognition.