【作者】 尚吉扬； 陈驰； 王文生；

【Author】 Shang Jiyang Chen Chi Wang Wensheng (Changchun University of Science and Technology,Laboratory of Contemporary Optical Measure Technology,Changchun,130022, China)

【机构】 长春理工大学现代光学测试技术研究室；

【摘要】 对畸变(比例缩放、旋转等)目标图像及用来合成最大平均相关高度滤波器的训练图像进行Sobel边缘提取,以增强畸变目标图像和滤波器的低频信息,再用最大平均相关高度频域滤波器对边缘增强后的畸变目标图像进行滤波,从而扩大了最大平均相关高度滤波器的畸变容差。通过这种将空域边缘增强和频域滤波增强相结合的方法,消除畸变对图像匹配识别的影响。更多还原

【Abstract】 The distorted(scale and rotation etc) target images and the training images are edge extracted by Sobel operator.The training images are used to synthesize Maximum Average Correlation Height filter.The low frequency information of the distorted target images and the filter is enhanced.Then the Maximum Average correlation Height filter is used to filter the edge extracted distorted target image.Thereby,the distortion tolerance of the Maximum Average Correlation Height filter is expanded.By this method,which the airspace edge extraction combines with frequency domain filtering,the influence of the distortion for the image pattern recognition is eliminated.更多还原