摘要：【中文】本发明公开一种训练数据生成方法，获取真实数据图像，所述真实数据图像包括手写数字和打印字体，所述手写数字和所述打印字体混合组成真实数据图像；切分所述真实数据图像中的手写数字图像，对所述切分的手写数字图像进行二值化处理后划分到对应数字的手写数字图像库中，对于所述打印字体在公开字体库匹配打印字体库；根据所述真实数据图像中的数据格式获取所述对应数字的手写数字图像库中对应的手写数字图像和所述打印字体库中的打印字体生成训练数据。根据本方法，通过制作更加真实的手写数字图像库，和选择打印自字体库，使生成真实度更高的训练数据，提高最终识别模型的准确率。 【EN】The invention discloses a training data generation method, which comprises the steps of obtaining a real data image, wherein the real data image comprises handwritten numbers and printing fonts, and the handwritten numbers and the printing fonts are mixed to form a real data image; segmenting the handwritten digital image in the real data image, dividing the segmented handwritten digital image into a handwritten digital image library of corresponding numbers after binarization processing is carried out on the segmented handwritten digital image, and matching the printing fonts with a printing font library in an open font library; and acquiring the corresponding handwritten digital image in the handwritten digital image library of the corresponding number and the printing font in the printing font library according to the data format in the real data image to generate training data. According to the method, by making a more real handwritten digital image library and selecting a self-printing font library, training data with higher reality degree is generated, and the accuracy of the final recognition model is improved.

摘要：【中文】本发明公开一种证件图像的截取方法，所述证件包括特定图标，所述特定图标可被图标识别模型识别；所示图标识别模型获取所述特定图标的特征值，依据所述特征值获得所述证件的倾斜角度和边长；根据所述证件的倾斜角度和边长截取所述证件图像。根据本发明提供的证件图像截取方法，利用图标识别模型对证件中的特定图标进行识别，依据特定图标的特征值实现对证件图像的矫正和截取，避免了获取大量真实数据训练识别模型，从根本上解决了使用个人数据的法律风险，同时可以降低开发成本简化开发过程，利用本方法截取证件图像过程简单，截取的证件图像准确率高。 【EN】The invention discloses a method for intercepting a certificate image, wherein the certificate comprises a specific icon which can be identified by an icon identification model; the icon identification model acquires a characteristic value of the specific icon, and acquires the inclination angle and the side length of the certificate according to the characteristic value; and intercepting the certificate image according to the inclination angle and the side length of the certificate. According to the certificate image intercepting method provided by the invention, the specific icon in the certificate is identified by the icon identification model, and the certificate image is corrected and intercepted according to the characteristic value of the specific icon, so that a large amount of real data is prevented from being acquired, the identification model is trained, the legal risk of using personal data is fundamentally solved, the development cost can be reduced, the development process can be simplified, the certificate image intercepting process is simple by the method, and the intercepted certificate image is high in accuracy.

摘要：【中文】本发明公开了一种图像文字的定位方法和装置，涉及计算机技术领域。该方法的一具体实施方式包括：利用文字定位模型获取目标图像中全部文字的位置信息；根据目标文字识别要素在目标图像中的位置特征，从全部文字的位置信息中确定目标文字识别要素对应的位置信息；根据目标文字识别要素分别在目标图像和模板图像中对应的位置信息和目标图像与模板图像之间的线性关系确定目标图像与模板图像之间的偏差系数；根据偏差系数和全部文字识别要素在模板图像中的位置信息确定每一个文字识别要素在目标图像中的位置信息。该实施方式降低了对模板图像的要求，提高了文字定位的准确率，提高了后续文字识别效率和文字识别效果。 【EN】The invention discloses a method and a device for positioning image characters, and relates to the technical field of computers. One embodiment of the method comprises: acquiring position information of all characters in a target image by using a character positioning model; according to the position characteristics of the target character recognition element in the target image, determining the position information corresponding to the target character recognition element from the position information of all characters; determining a deviation coefficient between the target image and the template image according to the corresponding position information of the target character recognition element in the target image and the template image and the linear relation between the target image and the template image; and determining the position information of each character recognition element in the target image according to the deviation coefficient and the position information of all the character recognition elements in the template image. The method and the device reduce the requirements on the template image, improve the accuracy of character positioning, and improve the subsequent character recognition efficiency and character recognition effect.

摘要：【中文】本申请提供了一种证件图像的识别方法，包括获取所述证件图像，基于文字定位模型对所述证件图像中文字所在的区域切片获得行切片图像和行切片图像的位置信息；根据所述位置信息对所述行切片图像分组；根据所述证件图像中的字段关键词按行出现的顺序，依次与行切片图像分组按顺序对应；基于文字识别模型对分组的所述行切片图像进行文字识别，获得每一组所述行切片图像中的文字；根据所述位置信息、字段关键词和字段关键词与所述分组的行切片图像对应关系输出所述证件图像的结构化文字信息。使用本方法获得的证件图像识别的结构化文字信息结果更加准确，可减少证件图像识别应用中的后期人工操作，提升工作效率。 【EN】The application provides a certificate image identification method, which comprises the steps of obtaining a certificate image, and slicing an area where characters are located in the certificate image based on a character positioning model to obtain a line slice image and position information of the line slice image; grouping the line slice images according to the position information; sequentially corresponding to the line slice image groups in sequence according to the sequence of the field keywords in the certificate image in the line appearance sequence; performing character recognition on the grouped line slice images based on a character recognition model to obtain characters in each group of line slice images; and outputting the structured character information of the certificate image according to the position information, the field keywords and the corresponding relation between the field keywords and the grouped line cutting images. The structured character information result of certificate image recognition obtained by the method is more accurate, the later manual operation in certificate image recognition application can be reduced, and the work efficiency is improved.