微型机与应用
微型機與應用
미형궤여응용
Microcomputer & its Applications
2015年
21期
40-42
,共3页
边缘检测%亚像素%Sobel%正交多项式%曲线拟合
邊緣檢測%亞像素%Sobel%正交多項式%麯線擬閤
변연검측%아상소%Sobel%정교다항식%곡선의합
edge detection%sub-pixel%Sobel%orthogonal polynomial%curve fitting
为了提高刀具预调测量仪的检测精度，提出了一种改进的图像快速亚像素边缘检测算法———基于正交多项式拟合的亚像素边缘检测算法。首先，利用传统的 Sobel 算子完成边缘点整像素级别的检测，确定边缘的主体区域；然后，过边缘点沿边缘法线方向拓展像素，取一系列像素点并计算其灰度值；最后根据像素点灰度分布的数学特征，利用正交多项式和最小二乘法求拟合函数，通过拟合曲线确定图像边缘点的精确位置，实现图像亚像素边缘检测。实验证明，该算法运行时间短，约为0.63 s；检测精度高，可达0.1 pixels。
為瞭提高刀具預調測量儀的檢測精度，提齣瞭一種改進的圖像快速亞像素邊緣檢測算法———基于正交多項式擬閤的亞像素邊緣檢測算法。首先，利用傳統的 Sobel 算子完成邊緣點整像素級彆的檢測，確定邊緣的主體區域；然後，過邊緣點沿邊緣法線方嚮拓展像素，取一繫列像素點併計算其灰度值；最後根據像素點灰度分佈的數學特徵，利用正交多項式和最小二乘法求擬閤函數，通過擬閤麯線確定圖像邊緣點的精確位置，實現圖像亞像素邊緣檢測。實驗證明，該算法運行時間短，約為0.63 s；檢測精度高，可達0.1 pixels。
위료제고도구예조측량의적검측정도，제출료일충개진적도상쾌속아상소변연검측산법———기우정교다항식의합적아상소변연검측산법。수선，이용전통적 Sobel 산자완성변연점정상소급별적검측，학정변연적주체구역；연후，과변연점연변연법선방향탁전상소，취일계렬상소점병계산기회도치；최후근거상소점회도분포적수학특정，이용정교다항식화최소이승법구의합함수，통과의합곡선학정도상변연점적정학위치，실현도상아상소변연검측。실험증명，해산법운행시간단，약위0.63 s；검측정도고，가체0.1 pixels。
In order to improve the detection accuracy of tool presetting and measuring machine , an improved sub-pixel edge detection algorithm for images is proposed. Firstly, the Sobel algorithm is carried out to get some edge points at pixel level, and then find the approximate position of edge. Secondly, along the normal of the edge direction with edge points, the pixel is expanded and its gray is calculated. According to mathematical feature of gray distribution of points, the gray level curve is fitted through orthogonal polynomial and least square fit method and the accurate position of edge point with a sub-pixel precision is achieved. The simulated experiments results of MATLAB show that the proposed novel algorithm needs much shorter running time, about 0.63 s, and much higher precision, about 0.1 pixels.
