Document OCR

下面有中文翻译 Hello everyone, My name is Dries and I'm working for the Belgian embassy in China. First I'll shortly explain about the services we offer at the embassy before going on to the project I would like to build with you. So, our service process is really simple: for example, a Belgian marries in China, so we will need to register his marriage in Belgium as well. Therefore we need an official translation of his/her marriage certificate. The married person takes the Chinese marriage certificate to a Chinese notary and they will provide an official translation which we can then use at the embassy. Often these notaries work together with people who don't have the necessary command of the language, so translations are low quality, but still expensive! About 90% of the text on these documents is always exactly the same, only name, age, sex, ... will be different, so here's what I would like to do with you: Build a website or program that recognizes which kind of document needs to be translated, so then it can just use the translation provided by myself to put the standard text into a file. The information that is variable (name, age, sex,...) will have to be recognized by the program so that it can also put this into the file. Don't worry about the translation part as this will be provided by myself. I don't know if I explained this clearly so if you have any questions please contact me. Thanks! Dries 大家好 ， 我叫Dries。我是比利时驻华大使馆的员工。首先我给你介绍一下我们大使馆提供的服务，再解释清楚想和你做哪样的项目。 大使馆的服务过程很简单：例如，一位比利时人在中国结过婚，所以我们得把他的婚姻也在比利时登记。因此，比利时大使馆需要中国当地的公证处提供的正式翻译。因为公证处往往和当地的学生合作（不和以荷兰文或法文为母语的人合作 ），所以公文的翻译也相当不准确，不过还是很贵！ 这些公文90%的信息基本上每次都一样，只有名字、年龄、性别等等会变得，所以我想和你设计一个识别公文的种类的软件。如此，软件用的是我提供的准确翻译再加上软件识别出来的可变信息（名字、年龄、性别等等），然后把所有的信息放在文档里面。这件文档就算是翻译好的公文。翻译部分都是我负责，请放心。 不知道我这样解释得清楚，有疑问的话随时联系我。 谢谢 Dries