CN115909340A - A method and device for collecting intelligent inspection certificates on the WEB side - Google Patents
A method and device for collecting intelligent inspection certificates on the WEB sideDownload PDFInfo
- Publication number
- CN115909340A CN115909340ACN202211449645.4ACN202211449645ACN115909340ACN 115909340 ACN115909340 ACN 115909340ACN 202211449645 ACN202211449645 ACN 202211449645ACN 115909340 ACN115909340 ACN 115909340A
- Authority
- CN
- China
- Prior art keywords
- certificate
- training
- photo
- license
- column
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Description
Translated fromChinese技术领域technical field
本发明涉及证照安全技术，尤其涉及一种WEB端智能检验证照采集的方法及装置。The invention relates to certificate security technology, in particular to a method and device for intelligent inspection and certificate collection at the WEB end.
背景技术Background technique
在证照管理领域，目前市面上主流的做法主要分为两种，一是：实时采集用户的证照，要求用户实时调起摄像头，进行拍照上传，系统调起摄像头时，会显示一个提示框，提示证照刚好占满提示框，用户拍摄后，上传到系统服务器，系统服务器通过使用OCR技术对证照进行识别，提取证照的关键信息，进一步核对信息是否有效。这种方式存在很大的缺陷，1、要求用户实时拍摄，不支持上传已有的证照，造成用户使用不便；2、系统显示的提示框，只是提示作用，并没有进行校验，用户可以随意拍摄证照，而系统并没有校验证照是否刚好占满提示框这个条件。二是：要求用户上传已存在的证照至系统服务器，系统服务器通过使用OCR技术对证照进行识别，提取证照的关键信息，进一步核对信息是否有效。In the field of license management, there are currently two mainstream methods on the market. One is: collect the user’s license in real time, and require the user to activate the camera in real time to upload photos. When the system activates the camera, a prompt box will be displayed, prompting The certificate just fills up the prompt box. After the user takes the photo, upload it to the system server. The system server uses OCR technology to identify the certificate, extract the key information of the certificate, and further check whether the information is valid. There are great defects in this method. 1. It requires users to take pictures in real time, and does not support uploading existing certificates, which causes inconvenience for users; 2. The prompt box displayed by the system is only a reminder and does not perform verification. Users can freely Take a photo of the certificate, but the system does not check whether the photo just fills the prompt box. The second is: the user is required to upload the existing certificate to the system server, and the system server uses OCR technology to identify the certificate, extract the key information of the certificate, and further check whether the information is valid.
由此可见，上述两种方式均存在较为明细的缺陷-没有校验证照是否满足证照管理要求，增加了人工审核校验环节。上述两种方式只要解决了证照的关键信息的提取和校验，并没有解决证照是否满足证照管理系统的诉求，例如：证照在照片中占比过大过小，证照清晰度是否满足条件、证照的是否倾斜、证照来源于复印件、翻拍。It can be seen that both of the above two methods have relatively detailed defects-there is no verification of whether the certificate meets the requirements of certificate management, and the manual review and verification process is added. The above two methods only solve the extraction and verification of the key information of the certificate, but do not solve whether the certificate meets the requirements of the certificate management system, for example: the proportion of the certificate in the photo is too large or too small, whether the clarity of the certificate meets the conditions, Whether it is tilted, the certificate comes from a copy, or a remake.
所述证照是包含证件内容的照片。The certificate is a photo containing the content of the certificate.
发明内容Contents of the invention
本发明目的在于提供一种WEB端智能检验证照采集的方法及装置，以解决上述现有技术存在的问题。The purpose of the present invention is to provide a method and device for collecting intelligent inspection certificates on the WEB side, so as to solve the above-mentioned problems in the prior art.
本发明中所述一种WEB端智能检验证照采集的方法包括以下三个阶段：A kind of WEB end intelligent inspection certificate collection method described in the present invention comprises the following three stages:
训练阶段：训练库采集训练样本；服务器利用OCR识别算法和OpenCV识别算法对训练样本进行识别然后存储在所述训练库；训练库对训练样本进行面积占比训练、清晰度训练、证照摆放斜率训练以及进行关键信息有效性的检验，统计训练后的数据，确定评分基准值，根据所述评分基准值确定评分策略，完成训练阶段；Training phase: the training library collects training samples; the server uses the OCR recognition algorithm and the OpenCV recognition algorithm to identify the training samples and then stores them in the training library; the training library performs area ratio training, clarity training, and license placement slope for the training samples Training and checking the validity of key information, counting the data after training, determining the scoring benchmark value, determining the scoring strategy according to the scoring benchmark value, and completing the training phase;
采集阶段：用户通过用户端选择或实时拍摄证照，通过web端上传至所述服务器；服务器利用OCR识别算法和OpenCV识别算法对证照进行信息提取并发送至训练库；训练库对证照进行面积占比计算、清晰度计算、证照摆放斜率计算以及进行关键信息有效性的检验，将结果返回至服务器；服务器得出多维度计算值；Acquisition stage: the user selects or takes a real-time photo of the certificate through the client terminal, and uploads it to the server through the web; the server uses the OCR recognition algorithm and the OpenCV recognition algorithm to extract information from the certificate and sends it to the training database; the training database calculates the area ratio of the certificate Calculation, clarity calculation, slope calculation of license placement and verification of the validity of key information, and return the results to the server; the server obtains multi-dimensional calculation values;
校验阶段：服务器调取训练库的评分基准值和评分策略对所述多维度计算值进行评分阈值检验，若评分低于阈值则通过web端向用户端提示不通过原因并提示用户端重新操作，若评分不低于阈值则对证照评分进行有效性检测；若有效性检测不通过则通过web端向用户端提示不通过原因并提示用户端重新操作，若有效性检测通过则通过web端向用户端提示证照采集完成。Verification stage: the server calls the scoring reference value and scoring strategy of the training library to check the scoring threshold of the multi-dimensional calculation value. If the score is lower than the threshold, it will prompt the user to fail the reason through the web terminal and prompt the user to re-operate , if the score is not lower than the threshold, the validity test will be performed on the license score; if the validity test fails, the reason for the failure will be prompted to the user terminal through the web terminal and the user terminal will be prompted to re-operate; if the validity test passes, the web terminal will be sent to The client prompts that the certificate collection is complete.
面积占比为证件中的头像栏、出生年月栏、性别栏、公民身份证号码栏其中一种的面积或多种的面积与照片面积的比值。The area ratio is the ratio of the area of one or more of the avatar column, date of birth column, gender column, and citizen ID card number column in the certificate to the area of the photo.
在面积占比训练和面积占比计算中，还判断证件距离比例信息；In the area ratio training and area ratio calculation, the certificate distance ratio information is also judged;
照片的垂直距离比例为：R(垂直)＝(y-0)/H；The vertical distance ratio of the photo is: R(vertical)=(y-0)/H;
照片的水平距离比例为：R(水平)＝(x-0)/W；The horizontal distance ratio of the photo is: R(horizontal)=(x-0)/W;
照片左上角为坐标原点，人头像栏左上角坐标为P(x,y)，H为照片高度，W为照片宽度。The upper left corner of the photo is the origin of the coordinates, the coordinates of the upper left corner of the avatar column are P(x,y), H is the height of the photo, and W is the width of the photo.
当R(垂直)大于基准值时判断证件水平偏下，当R(垂直)小于基准值时判断证件水平偏上；R(水平)大于基准值时判断证件水平偏右，当R(水平)小于基准值时判断证件水平偏左。When R (vertical) is greater than the reference value, it is judged that the certificate is horizontally downward; when R (vertical) is smaller than the reference value, it is judged that the certificate is horizontal; when R (horizontal) is greater than the reference value, it is judged that the certificate is horizontally right; When the reference value is used, it is judged that the level of the certificate is to the left.
证照摆放斜率具体为：The slope of the license placement is specifically:
k＝|(y6-y2)/(x6-x2)|；k=|(y6-y2)/(x6-x2)|;
姓名栏左上角的坐标为(x2,y2)，公民身份证号码栏左上角的坐标为(x6,y6)；The coordinates of the upper left corner of the name column are (x2, y2), and the coordinates of the upper left corner of the citizen ID card number column are (x6, y6);
当x6不等于x2时计算证照摆放斜率k，k越大表示照片倾斜度越大，k越小表示照片倾斜度越低；When x6 is not equal to x2, calculate the slope k of the certificate placement. The larger the k, the larger the slope of the photo, and the smaller the k, the lower the slope of the photo;
当x6等于x2时，判断证件没有倾斜。When x6 is equal to x2, it is judged that the certificate is not tilted.
本发明所述一种WEB端智能检验证照采集的装置，利用所述方法进行证照采集和检验。According to the present invention, a WEB-end intelligent inspection certificate collection device utilizes the method for certificate collection and inspection.
本发明中所述一种WEB端智能检验证照采集的方法及装置，其优点在于，基于OCR技术和OpenCV技术识别、提取证照的关键信息、以及照片的属性信息。通过大量的证照数据进行模型训练，得到一组满足需求的评分基准值，服务器对用户上传的证照进行OCR识别、提取数据、进行评分，得到的评分值与AI模型评分基准值进行比较，根据比较结果判断证照是否有效。引入了系统评分机制，通过多维度的统计分析数据，有效解决了证照占比过大过小、证照模糊、证照倾斜摆放、证照复印件、证照翻拍的问题，大大提高了证照有效率，减轻了审查工作人员的工作，节约了审核工作人员、系统用户的宝贵时间，提高了工作效率。The method and device for intelligent inspection and certificate collection at the WEB end described in the present invention have the advantage of identifying and extracting key information of certificates and attribute information of photos based on OCR technology and OpenCV technology. Through a large amount of certificate data for model training, a set of scoring reference values that meet the requirements are obtained. The server performs OCR recognition, data extraction, and scoring on the certificates uploaded by users. The obtained scoring values are compared with the AI model scoring reference values. According to the comparison The results determine whether the certificate is valid. The introduction of a system scoring mechanism, through multi-dimensional statistical analysis of data, effectively solves the problems of too large or too small proportion of certificates, blurred certificates, oblique placement of certificates, photocopies of certificates, and copying of certificates, which greatly improves the efficiency of certificates and reduces It saves the work of review staff, saves valuable time of review staff and system users, and improves work efficiency.
附图说明Description of drawings
图1是本发明中所述一种WEB端智能检验证照采集的方法流程图。Fig. 1 is a flow chart of a method for collecting intelligent inspection certificates at the WEB end described in the present invention.
具体实施方式Detailed ways
如图1所示，本发明中所述一种WEB端智能检验证照采集的装置包括用户端、web端、服务器和训练库，其中web端可以是浏览器。用户在用户端通过浏览器与服务器进行通讯。训练库用于通过大量的训练样本进行训练得到AI模型，以及后续利用AI模型进行计算得到多维度计算值。服务器主要用于AI模型的调度和检验判断。各功能模块具体工作流程即为本发明所述一种WEB端智能检验证照采集的方法。As shown in FIG. 1 , a device for collecting intelligent verification certificates at the WEB terminal in the present invention includes a client terminal, a web terminal, a server and a training library, wherein the web terminal may be a browser. The user communicates with the server through the browser on the client side. The training library is used to obtain the AI model through training with a large number of training samples, and then use the AI model to perform calculations to obtain multi-dimensional calculation values. The server is mainly used for the scheduling and inspection and judgment of the AI model. The specific work flow of each functional module is a method for collecting intelligent inspection certificates at the WEB end of the present invention.
(一)模型训练阶段的步骤(1) Steps in the model training phase
1.开始训练指定证照模型，在本实施例中以以身份证证照为例。本领域技术人员根据公知常识可知，证照还可以包括驾驶证、护照、户口簿、律师证等等。1. Start training the specified certificate model, in this embodiment, take the ID card as an example. Those skilled in the art can know according to the common knowledge that the certificate can also include a driver's license, a passport, a residence booklet, a lawyer's certificate, and the like.
2.采集足够多的证照原始数据，为了更好训练数据模型，随机采集足够多的证照原始样例。样例应该包括证照占比过大过小、证照模糊、证照倾斜摆放、证照复印件、证照翻拍等实际应用可能出现的情况样例。2. Collect enough original license data. In order to better train the data model, randomly collect enough original license samples. The samples should include examples of situations that may occur in practical applications such as the proportion of the certificate being too large or too small, the certificate being blurred, the certificate being placed at an angle, the copy of the certificate, and the photo of the certificate.
3.使用OCR技术对证照进行识别和提取关键信息，采集到标准格式的证照。3. Use OCR technology to identify and extract key information from licenses, and collect licenses in standard formats.
OCR识别并提取关键信息，所述关键信息包括但不限于：姓名、性别、民族、出生年月、住址、身份证号码、人头像各模块，以及所述各模块的左上角的点坐标、长高，整体照片的长高等。OCR identifies and extracts key information, which includes but is not limited to: name, gender, ethnicity, date of birth, address, ID number, head portrait modules, and the point coordinates and length of the upper left corner of each module. Height, the length and height of the overall photo, etc.
提取到的关键信息如下：The key information extracted is as follows:
照片的左上角坐标O(0,0)，相当于坐标原点，照片的长度：W，照片的高度：H，证件的长度：w，证件的高度：h。The coordinate O(0,0) of the upper left corner of the photo is equivalent to the origin of the coordinates, the length of the photo: W, the height of the photo: H, the length of the certificate: w, the height of the certificate: h.
人头像栏的左上角坐标P(x,y)，人头像栏的长度：w1，人头像栏的高度：h1。The coordinates P(x,y) of the upper left corner of the portrait column, the length of the portrait column: w1, and the height of the portrait column: h1.
姓名栏的左上角坐标N(x2,y2)，姓名栏的长度：w2，姓名栏的高度：h2。The coordinates of the upper left corner of the name column are N(x2, y2), the length of the name column: w2, and the height of the name column: h2.
性别栏的左上角坐标S(x3,y3)，性别栏的长度：w3，性别栏的高度：h3。The coordinate S(x3,y3) of the upper left corner of the gender column, the length of the gender column: w3, and the height of the gender column: h3.
民族栏的左上角坐标Na(x4,y4)，民族栏的长度：w4，民族栏的高度：h4。Coordinate Na(x4,y4) of the upper left corner of the ethnicity column, the length of the ethnicity column: w4, the height of the ethnicity column: h4.
住址栏的左上角坐标D(x5,y5)，住址栏的长度：w5，住址栏的高度：h5。The coordinates of the upper left corner of the address bar are D(x5, y5), the length of the address bar: w5, and the height of the address bar: h5.
公民身份证号码栏的左上角坐标I(x6,y6)，公民身份证号码栏的长度：w6，公民身份证号码栏的高度：h6。Coordinate I(x6, y6) of the upper left corner of the ID card number column, the length of the ID card number column: w6, and the height of the ID card number column: h6.
4.校验关键信息是否正确，如果核对关键信息错误或者没有识别出关键信息，则返回步骤1；否则，进入步骤5。4. Check whether the key information is correct. If the key information is wrong or the key information is not identified, return to step 1; otherwise, go to step 5.
5.使用opencv技术对照片进行清晰度检测；利用拉普拉斯算子计算照片的二阶导数，反映照片的边缘信息，同样事物的照片，清晰度高的，相对应的经过拉普拉斯算子滤波后的照片的方差也就越大。通过大量的证照原始数据进行训练，统计、分析实验数据，根据整体照片质量确定清晰度阀值C。有了代表清晰度C值，剩下的工作就是设定相应的阀值，如果某照片方差低于预先定义的阈值C，那么该照片就可以被认为是模糊的，高于阈值C，就不是模糊的。5. Use opencv technology to detect the sharpness of the photo; use the Laplacian operator to calculate the second derivative of the photo to reflect the edge information of the photo, and the photos of the same thing, with high definition, correspond to Laplacian The variance of the photo filtered by the operator is larger. Through a large number of original license data for training, statistics and analysis of experimental data, the definition threshold C is determined according to the overall photo quality. With the C value representing the sharpness, the remaining work is to set the corresponding threshold. If the variance of a photo is lower than the predefined threshold C, then the photo can be considered blurred. If it is higher than the threshold C, it is not vague.
6.证件占比大小算法，在OCR识别的关键信息中，头像栏、出生年月栏、性别栏、公民身份证号码的面积占比比较稳定。相反，姓名栏、民族栏、住址的面积分别会随着姓名字数长度、民族字数长度、住址字数长度变化而变化，不适合被使用作为计算面积占比算法的基准元素。基于上述分析，头像栏、出生年月栏、性别栏、公民身份证号码均可以用来计算证件占比大小，实际应用可以根据训练结果，选择其一或者任意选择多种作为校验基准。6. Algorithm of the proportion of documents. Among the key information identified by OCR, the area proportions of the profile picture column, date of birth column, gender column, and citizen ID number are relatively stable. On the contrary, the area of the name column, ethnicity column, and address will change with the length of the word length of the name, ethnicity, and address, respectively, so it is not suitable to be used as a benchmark element for calculating the area ratio algorithm. Based on the above analysis, the avatar column, date of birth column, gender column, and citizen ID card number can all be used to calculate the proportion of the certificate. In practical applications, one or more of them can be selected as the verification benchmark according to the training results.
本实施例中，以人头像栏面积占照片总面积的比例作为为计算面积占比算法的元素。人头像栏面积S1＝w1*h1；照片面积S＝W*H；人头像占比R1＝S1/S2。令标准情况下，人头像占比基准值为R1`。判断R1和R1`的关系，R1过大表示证件没有完整处于照片中，R1过小表示证件处于照片中太小。In this embodiment, the ratio of the area of the head portrait column to the total area of the photo is used as an element of the algorithm for calculating the area ratio. Head portrait column area S1=w1*h1; photo area S=W*H; head portrait ratio R1=S1/S2. Under standard conditions, the base value of the proportion of human head portraits is R1`. Judge the relationship between R1 and R1`. If R1 is too large, it means that the certificate is not completely in the photo. If R1 is too small, it means that the certificate is too small in the photo.
7.检验证件所处位置是否合理，由步骤3得到各个元素的左上角坐标。人头像栏左上角与照片左上角的水平、垂直距离比例如下：7. Check whether the position of the verification document is reasonable, and obtain the coordinates of the upper left corner of each element from step 3. The ratio of the horizontal and vertical distances between the upper left corner of the profile picture bar and the upper left corner of the photo is as follows:
人头像栏左上角距离照片上边框的垂直距离比例：R(垂直)＝(y-0)/H；The ratio of the vertical distance between the upper left corner of the portrait column and the upper border of the photo: R (vertical) = (y-0)/H;
人头像栏左上角距离照片上边框的垂直距离比例：R(水平)＝(x-0)/W；The ratio of the vertical distance between the upper left corner of the portrait column and the upper frame of the photo: R(horizontal)=(x-0)/W;
通过足够多的训练数据，得到足够多的训练结果R(垂直)及R(水平)，对R(垂直)、R(水平)进行统计、分析，可以得到满足证照校验R(垂直)的基准值以及R(水平)的基准值。Through enough training data, enough training results R (vertical) and R (horizontal) can be obtained, and R (vertical) and R (horizontal) can be counted and analyzed to obtain a benchmark for certificate verification R (vertical). value and the reference value of R (horizontal).
当R(垂直)大于基准值时判断证件水平偏下，当R(垂直)小于基准值时判断证件水平偏上；R(水平)大于基准值时判断证件水平偏右，当R(水平)小于基准值时判断证件水平偏左。When R (vertical) is greater than the reference value, it is judged that the certificate is horizontally downward; when R (vertical) is smaller than the reference value, it is judged that the certificate is horizontal; when R (horizontal) is greater than the reference value, it is judged that the certificate is horizontally right; When the reference value is used, it is judged that the level of the certificate is to the left.
8.检验证件摆放是否倾斜：获取姓名栏和公民身份证号码栏的左上角坐标，当x6不等于x2时使用斜率公式(x6不等于x2)：k＝|(y6-y2)/(x6-x2)|。。斜率k绝对值越小，表示姓名栏和公民身份证号码栏的水平对齐偏差越大，反之，偏差越小。8. Check whether the certificate is placed tilted: obtain the coordinates of the upper left corner of the name column and the citizen ID number column, and use the slope formula when x6 is not equal to x2 (x6 is not equal to x2): k=|(y6-y2)/(x6 -x2)|. . The smaller the absolute value of the slope k, the larger the horizontal alignment deviation between the name column and the citizen ID card number column, and vice versa, the smaller the deviation.
当x6等于x2时，判断姓名栏和公民身份证号码栏在水平方向对齐。When x6 is equal to x2, it is judged that the name column and the citizen ID number column are aligned in the horizontal direction.
9.通过上述步骤训练AI模型，获得多项校验证件的基准值：9. Train the AI model through the above steps to obtain the benchmark values of multiple verification documents:
a)清晰度C基准值a) Clarity C reference value
b)人头像面积占比基准值R1`b) The reference value of the proportion of the head portrait area R1`
c)人头像水平、垂直距离比例基准值R(水平)、R(垂直)c) The horizontal and vertical distance ratio reference values R (horizontal) and R (vertical) of the head portrait
d)姓名栏和公民身份证号码栏的斜率k.d) The slope k of the name column and the citizen ID card number column.
10.根据各项基准值，引入评分策略机制，本领域技术人员可以在实际应用中根据需要再定义评分策略。在本实施例中，给出一个评分策略：10. According to each benchmark value, a scoring policy mechanism is introduced, and those skilled in the art can redefine the scoring policy according to needs in practical applications. In this example, a scoring strategy is given:
11.把评分策略保存到训练库。11. Save the scoring strategy to the training library.
12.结束训练。12. End training.
(二)证照采集阶段的步骤(2) Steps in the license collection stage
1.用户进入用户端的证照采集页面。1. The user enters the license collection page of the client.
2.用户选择上传已有证照或实时拍摄。2. The user chooses to upload an existing license or take a real-time photo.
3.用户上传步骤2的证照。3. The user uploads the certificate of step 2.
4.服务器进行OCR和OpenCV的识别、提取关键信息。4. The server performs OCR and OpenCV identification and extracts key information.
5.核验关键信息，如核验失败，跳转步骤2，否则，执行下一步操作。5. Verify the key information, if the verification fails, go to step 2, otherwise, go to the next step.
6.进行算法检测，得到各项检测结果值。6. Carry out algorithm detection to obtain the values of various detection results.
(三)智能核验阶段的步骤(3) Steps in the intelligent verification stage
1.获取用户上传证照的各项检测结果值。1. Obtain the test result values of the certificates uploaded by the user.
2.根据评分策略，对检测结果值进行评分、累计。2. According to the scoring strategy, score and accumulate the test result value.
3.得到该证照的最后得分值。3. Obtain the final score value of the certificate.
4.该证照的最后得分与评分策略预设有效证照的阈值进行比较，如果大于阈值，证照核验通过，保存用户有效证照；反之，提示用户重新上传证照，并提示原因。4. The final score of the certificate is compared with the threshold value of the valid certificate preset by the scoring strategy. If it is greater than the threshold value, the certificate verification is passed and the user's valid certificate is saved; otherwise, the user is prompted to re-upload the certificate and the reason is prompted.
对于本领域的技术人员来说，可根据以上描述的技术方案以及构思，做出其它各种相应的改变以及形变，而所有的这些改变以及形变都应该属于本发明权利要求的保护范围之内。For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and ideas described above, and all these changes and deformations should fall within the protection scope of the claims of the present invention.
Claims (6)
Translated fromChinesePriority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211449645.4ACN115909340A (en) | 2022-11-18 | 2022-11-18 | A method and device for collecting intelligent inspection certificates on the WEB side |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211449645.4ACN115909340A (en) | 2022-11-18 | 2022-11-18 | A method and device for collecting intelligent inspection certificates on the WEB side |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115909340Atrue CN115909340A (en) | 2023-04-04 |
Family
ID=86496702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211449645.4APendingCN115909340A (en) | 2022-11-18 | 2022-11-18 | A method and device for collecting intelligent inspection certificates on the WEB side |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115909340A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107016608A (en)* | 2017-03-30 | 2017-08-04 | 广东微模式软件股份有限公司 | The long-range account-opening method and system of a kind of identity-based Information Authentication |
| CN114419008A (en)* | 2022-01-24 | 2022-04-29 | 北京译图智讯科技有限公司 | Image quality evaluation method and system |
| WO2022121039A1 (en)* | 2020-12-10 | 2022-06-16 | 广州广电运通金融电子股份有限公司 | Bankcard tilt correction-based detection method and apparatus, readable storage medium, and terminal |
| CN114820476A (en)* | 2022-04-12 | 2022-07-29 | 中科计算技术创新研究院 | Identification card identification method based on compliance detection |
| CN114926829A (en)* | 2022-05-30 | 2022-08-19 | 成都新希望金融信息有限公司 | Certificate detection method and device, electronic equipment and storage medium |
- 2022
- 2022-11-18CNCN202211449645.4Apatent/CN115909340A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107016608A (en)* | 2017-03-30 | 2017-08-04 | 广东微模式软件股份有限公司 | The long-range account-opening method and system of a kind of identity-based Information Authentication |
| WO2022121039A1 (en)* | 2020-12-10 | 2022-06-16 | 广州广电运通金融电子股份有限公司 | Bankcard tilt correction-based detection method and apparatus, readable storage medium, and terminal |
| CN114419008A (en)* | 2022-01-24 | 2022-04-29 | 北京译图智讯科技有限公司 | Image quality evaluation method and system |
| CN114820476A (en)* | 2022-04-12 | 2022-07-29 | 中科计算技术创新研究院 | Identification card identification method based on compliance detection |
| CN114926829A (en)* | 2022-05-30 | 2022-08-19 | 成都新希望金融信息有限公司 | Certificate detection method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8284985B2 (en) | Interactive device for processing documents | |
| CN108241836A (en) | Method and device for security inspection | |
| CN102087800A (en) | Digital online paper-marking system | |
| CN105320952B (en) | A kind of driving license information identifying method based on OCR | |
| CN111784498A (en) | Identity authentication method, device, electronic device and storage medium | |
| CN109446345A (en) | Nuclear power file verification processing method and system | |
| CN114820476B (en) | ID card recognition method based on compliance detection | |
| CN116798061A (en) | Bill auditing and identifying method, device, terminal and storage medium | |
| CN105469513A (en) | Self-service all-in-one machine based on face detection and character recognition and using method thereof | |
| CN112487982A (en) | Merchant information auditing method, system and storage medium | |
| CN117392790A (en) | A method, system and equipment for borrowing and returning books based on visual recognition | |
| US11527102B2 (en) | Systems and methods of automated biometric identification reporting | |
| CN116311299A (en) | Method, device and system for identifying structured data of table | |
| CN114596439A (en) | Image-based intelligent damage assessment and check method and device for car insurance and electronic equipment | |
| CN114821725A (en) | A Miner Face Recognition System Based on Neural Network | |
| CN115909340A (en) | A method and device for collecting intelligent inspection certificates on the WEB side | |
| CN205563854U (en) | Serve all -in -one by oneself based on face detection and character recognition | |
| CN111401438B (en) | Image sorting method, device and system | |
| TWI777188B (en) | Contract signature authentication method and device | |
| CN115690819A (en) | Big data-based identification method and system | |
| KR101429929B1 (en) | Biometric and demographic data transmission and management systems and devices | |
| CN111967887A (en) | Remote appraising method and system for digital stamps and coins and computer-readable storage medium | |
| WO2021073576A1 (en) | Intelligent vehicle verification method and system, computer device, and storage medium | |
| CN120639506B (en) | AI identity authentication system and method based on user behavior | |
| TWI879701B (en) | Purchase and sales and other voucher identification system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |