CAJ | 학술논문

针对复杂背景的视频图像车型识别，提出了一种利用尺度显著性的车型识别方法。由于尺度显著性对图像均一亮度变化、缩放、旋转以及噪声都具有不变性，因此引入尺度显著性算法提取车辆图像的分类特征。最后采用RBF网络分类验证该方法对多种车型的识别。实验结果表明,提取尺度显著性特征能够有效地识别汽车车型。
침대복잡배경적시빈도상차형식별，제출료일충이용척도현저성적차형식별방법。유우척도현저성대도상균일량도변화、축방、선전이급조성도구유불변성，인차인입척도현저성산법제취차량도상적분류특정。최후채용RBF망락분류험증해방법대다충차형적식별。실험결과표명,제취척도현저성특정능구유효지식별기차차형。
In light of the complicate background of vehicle image in vehicle identification system, this paper proposes a method of vehicle type recognition based on scale saliency algorithm. Because the image features extracted by scale saliency algorithm is invariant on image uniform brightness variations, scaling change, rotation and noise, the scale saliency algorithm is introduced to extract classification features to various vehicle types. At last, the RBF neural network is constructed to verify the feasibility of the proposed method. Experimental result show that the vehicle types are recognized effectively and accurately by the method of scale salient features.