针对复杂场景中车辆由于视角变化引起的检测精确度过低的问题,改进霍夫投票目标检测模型,提出一种在统一框架下通过不同权重组合发现目标最优视角并进行精确定位的方法。首先,利用一种无监督方法实现多视角车辆的子视角划分;其次,利用...针对复杂场景中车辆由于视角变化引起的检测精确度过低的问题,改进霍夫投票目标检测模型,提出一种在统一框架下通过不同权重组合发现目标最优视角并进行精确定位的方法。首先,利用一种无监督方法实现多视角车辆的子视角划分;其次,利用子视角划分结果定义霍夫投票过程中各正例样本在不同视角下的投票权重;最后,利用子视角划分和投票权重,提出一种新的适用于多视角目标检测的加权霍夫投票模型。在MITStreet Scene Cars和PASCAL VOC2007 Cars两个常用数据集上的实验结果表明,所提方法在不增加模型复杂度的前提下,有效提升了多视角目标检测精确度。展开更多
Corner detection is a chief step in computer vision. A new corner detection algorithm in planar curves is proposed. Firstly, from the human perception, two key characteristics are given as an amendment of the traditio...Corner detection is a chief step in computer vision. A new corner detection algorithm in planar curves is proposed. Firstly, from the human perception, two key characteristics are given as an amendment of the traditional corner properties. Based on the two properties, the concept of the fuzzy set is introduced into a detection. Secondly, the extracted-formulae of three groups including the features of the corner subject degree are derived. Through synthesizing the features of three groups, the judgments of the corner detection, location, and optimization are obtained. Finally, by using the algorithm the detection results of several examples and feature curves for some interested parts, as well as the detection results for the test images history in references are given. Results show that the algorithm is easily realized after adopting the fuzzy set, and the detection effect is very ideal.展开更多
In manufacture of precise optical products, it is important to inspect and classify the potential defects existing on the products’ surfaces after precise machining in order to obtain high quality in both functionali...In manufacture of precise optical products, it is important to inspect and classify the potential defects existing on the products’ surfaces after precise machining in order to obtain high quality in both functionality and aesthetics. The existing methods for detecting and classifying defects all are low accuracy or efficiency or high cost in inspection process. In this paper, a new inspection system based on machine vision has been introduced, which uses automatic focusing and image mosaic technologies to rapidly acquire distinct surface image, and employs Case-Based Reasoning(CBR)method in defects classification. A modificatory fuzzy similarity algorithm in CBR has been adopted for more quick and robust need of pattern recognition in practice inspection. Experiments show that the system can inspect surface diameter of 500mm in half an hour with resolving power of 0.8μm diameter according to digs or 0.5μm transverse width according to scratches. The proposed inspection principles and methods not only have meet manufacturing requirements of precise optical products, but also have great potential applications in other fields of precise surface inspection.展开更多
文摘针对复杂场景中车辆由于视角变化引起的检测精确度过低的问题,改进霍夫投票目标检测模型,提出一种在统一框架下通过不同权重组合发现目标最优视角并进行精确定位的方法。首先,利用一种无监督方法实现多视角车辆的子视角划分;其次,利用子视角划分结果定义霍夫投票过程中各正例样本在不同视角下的投票权重;最后,利用子视角划分和投票权重,提出一种新的适用于多视角目标检测的加权霍夫投票模型。在MITStreet Scene Cars和PASCAL VOC2007 Cars两个常用数据集上的实验结果表明,所提方法在不增加模型复杂度的前提下,有效提升了多视角目标检测精确度。
文摘Corner detection is a chief step in computer vision. A new corner detection algorithm in planar curves is proposed. Firstly, from the human perception, two key characteristics are given as an amendment of the traditional corner properties. Based on the two properties, the concept of the fuzzy set is introduced into a detection. Secondly, the extracted-formulae of three groups including the features of the corner subject degree are derived. Through synthesizing the features of three groups, the judgments of the corner detection, location, and optimization are obtained. Finally, by using the algorithm the detection results of several examples and feature curves for some interested parts, as well as the detection results for the test images history in references are given. Results show that the algorithm is easily realized after adopting the fuzzy set, and the detection effect is very ideal.
文摘In manufacture of precise optical products, it is important to inspect and classify the potential defects existing on the products’ surfaces after precise machining in order to obtain high quality in both functionality and aesthetics. The existing methods for detecting and classifying defects all are low accuracy or efficiency or high cost in inspection process. In this paper, a new inspection system based on machine vision has been introduced, which uses automatic focusing and image mosaic technologies to rapidly acquire distinct surface image, and employs Case-Based Reasoning(CBR)method in defects classification. A modificatory fuzzy similarity algorithm in CBR has been adopted for more quick and robust need of pattern recognition in practice inspection. Experiments show that the system can inspect surface diameter of 500mm in half an hour with resolving power of 0.8μm diameter according to digs or 0.5μm transverse width according to scratches. The proposed inspection principles and methods not only have meet manufacturing requirements of precise optical products, but also have great potential applications in other fields of precise surface inspection.
