摘要
用于车型轮廓识别的汽车图片通常带有极复杂的背景噪音。在此,采用等高线方法提取车型轮廓并与传统图像处理方法(如:Canny算子、形态学算子)进行了比较,实验结果表明:等高线原理应用于复杂背景情况下车型轮廓的提取效果更为理想。解决了在单一背景下提取轮廓线的局限性问题。并用直线拟合车型轮廓曲线,根据不同的车型会有不同直线斜率和成角信息,提出了斜率之差算法,用于等高线轮廓的直线拟合。该算法为后续车型分类识别提供了非常有用和可靠的特征样本信息。
Generally speaking, there is complex background noise in the photographs used to recognize vehicles. The principle of contour is used to get the contours of vehicles. A comparison is made between the results of traditional methods of digital image processing （for example： Canny operator, morphologic operator） and the results of using the principle of contour in the experimentation. It is shown that the principle of contour used to get the vehicle contours would gain more perfectly effect. The limitation to get the contour of vehicle in the single background is solved by author. Based on different slope and angle information on different contours of vehicles, the arithmetic is brought forward into fitting beeline with difference of slope, which would provide usable and credible information to the later distinguishing of the vehicle contours.
出处
《计算机工程与设计》
CSCD
北大核心
2008年第13期3496-3498,共3页
Computer Engineering and Design
关键词
等高线
车型轮廓线
斜率之差
直线拟合
车型识别
contour line
contours of vehicle
difference of slope
fitting beeline
recognition of vehicle
