图像边缘特征提取算法设计与研究

Design and Research of Image Edge Feature Extraction Algorithm

随着人工智能技术的快速发展,边缘特征提取是图像数据处理中的重要一环,图像数据的边缘特征数据提取的完整性及准确性关系到图像应用的效果。现有的图像边缘特征提取算法在提取特征数据时,存在边缘轮廓特征不清晰及提取耗时长等问题,尤其针对存在遮挡或复杂背景的图像时,存在边缘特征雾化,甚至出现图像目标特征轮廓缺失现象。针对上述问题,以图像数据的高质量边缘特征提取需求为牵引,结合图像数据边缘特征提取方法同步设计边缘特征提取流程,并开展图像数据边缘特征提取算法系统的图像滤波功能、图像平滑处理、图像边缘定位、阈值分割、边缘检测及边缘特征提取等功能单元设计,同时将提取的算法进行封装处理,便于各个功能组合应用。通过对功能单元算法及实验测试,该系统设计的图像边缘特征提取算法具有效率高,提取质量高且数量大等优点,能够满足复杂图像特征提取需求。

The existing image edge feature extraction algorithms have problems with unclear edge contour features and long extraction time when extracting feature data,especially for images with occluded or complex backgrounds,where edge feature fogging occurs and even the image target feature contour is missing.In response to the above issues,this paper takes the demand for high-quality edge feature extraction of image data as the driving force,and combines the edge feature extraction method of image data to synchronously design the edge feature extraction process.It also carries out the design of functional units such as image filtering function,image smoothing processing,image edge positioning,threshold segmentation,edge detection,and edge feature extraction in the image data edge feature extraction algorithm system,Simultaneously encapsulate the extracted algorithm for the convenience of combining various functions for application.