求专业人工英语翻译论文介绍了车型分类识别系统的发展现状,举出了几种目前国内外所采用的车型自动分类识别的方法,并分析了各自

求专业人工英语翻译
论文介绍了车型分类识别系统的发展现状,举出了几种目前国内外所采用的车型自动分类识别的方法,并分析了各自的优缺点；介绍了基于图像识别的车型分类识别系统的研究现状,归纳了车型分类识别系统的工作流程、核心问题和关键技术,评价了目前存在的车型分类识别标准车辆特征选取方法的不足之处.
通过对车型图像进行phog等全局特征的提取,再用支持向量机SVM进行分类.在第一阶段的分类过程中,将提取出来的全局特征向量作为SVM支持向量机的输入数据,然后得到车辆的形状分类结果；在第二阶段的分类过程中,分析实验结果,调整提取特征的参数,进一步完善实验.论文最后给出了实验结果和分析,最终分类正确率达到84％,验证了该车型分类系统的有效性.

1,This paper introduces the classification system models of the current situation of the development, cited several vehicles at home and abroad and the automatic classification identification method, and analyzes their advantages; This paper presents the image recognition model classification system, summarizes the present situation of the research models classification work flow of the system, the core problem and key technology, evaluated the existing models classification standard vehicle feature selection method shortcomings.
Through the of vehicle image characteristics such as the extraction of global phog, garnish with support vector machine SVM classification. In the first stage of the classification process, will the global properties of extracted) for SVM support vector machine (SVM) input data, then get the car shape classification results; In the second phase of the classification process, analyzing the results, and adjust the parameters of the extracted features, further perfecting experiment. The paper gives the experimental results and analysis, finally classification accuracy reached 84%, and verifies the effectiveness of the model classification system.
2,Paper describes the development of vehicle classification and identification system status, citing several models currently used by domestic and foreign auto-classification and identification methods and analysis of their strengths and weaknesses; introduced classification models based on image recognition identification system of Research, summarized the vehicle classification and identification system workflow, the core issues and key technologies, evaluation of the existing vehicle classification standard vehicle feature selection methods identify inadequacies. By models such as global phog image feature extraction, and then support vector machines SVM classification. In the first stage of the classification process, the global feature vector extracted from the SVM support vector machine as input data, then get the shape of the vehicle classification; in the second stage of the classification process, the analysis of experimental results, adjust the feature extraction parameters, further improve the experiment. Paper, the experimental results and the final analysis, the final classification accuracy rate of 84%, verified the validity of the vehicle classification system.