Data entry may be tedious and demanding, but there are ways to make sure you deliver.
The issue I am facing is the conversion of the Google OCR coordinates for the OCRd text the image dimensions within the application that are the same to the device width and height size. The dimensions required for Google Vision OCR is maximum of 1280 by height:width ratio. This is referred to as a document size. The image captured on device is equal to device height and width. The coordinates of the text blocks that I want come as a response of OCR request from Google in the document dimensions described above. I need to convert the coordinates to the device height and width dimensions to map the texts block so that a user can tap the screen and I can check if there is a book in that screen location. While at the same time drawing an overlay rect to show the region covered by the co...
I am going to implement ocr function to mobile. First, you have to extract all texts from image by mobile. App can be applied to image saved in mobile and camera. What is important is the accuracy. Inputs images have several style(light condition, skew, faint, etc) I have already made the app, but its accuracy is not high in skewed images. You have to resolve. Second, you have to apply ner model and DNN to get entity of certain in images. Third, you have to implement my important requirements. (I will share in chat) I want to implement all in mobile. you have to deploy in mobile. You have to be confidence in MLkit, tensorflow lite, android studio, java, ocr etc. If you don't have skills above, don't please bid. We must not to waste our time and effort. When you are only expert, b...
Top OCR Community Articles
Does maintaining accuracy in data entry for your company seem a daunting task for you? Here are a few tips to ease the pressure.
Are flaws in data entry eating up your revenues? Here are 11 ways to get rid of such issues and increase your company’s overall productivity.