Ocr Tesseract Jobs

3 were found based on your criteria {{ paging.total|number:0 }} were found based on your criteria

show all
  • Hourly ({{ jobTypeController.getFacetCount("0")|number:0}})
  • Fixed Price ({{ jobTypeController.getFacetCount("1")|number:0}})
Hourly - Intermediate ($$) - Est. Time: Less than 1 month, 10-30 hrs/week - Posted
Hi , We nee a Java lib( or jar file ) , that can read an image and extract specific text content from that, which will be returned as a JSON. Following are the detailed specs for this -The image can be of varying qualities( not bright enough, clicked from a mobile camera, min 5mp camera) but text to be extracted will always be typed text and not handrwitten -If the quality of the image is not right the first time, the application should be able to manipulate the image , eg. play around with contrast, sharpening, etc so as to make it more suitable for data extraction. -If the quality of the cannot be improved at all , such that the text cannot be extracted, then it should return a specific error code indicating the same. Please indicate what parameters would you use to determine this quality( eg. size of image, contrast ratio, etc ) -Your program can use a third party OCR softwares for the same, but these should be open source and not paid one. To finally test the same, I would be providing images which can be used as test case. Any other question, let me know.
Skills: OCR Tesseract Computer vision Java OCR algorithms
Hourly - Intermediate ($$) - Est. Time: 1 to 3 months, Less than 10 hrs/week - Posted
Following internal testing with Tesseract, we know understand that we need an experienced OCR person to help us train the OCR with a large number of Print publication. We want to extract the content from over 100 publication in UK, but also in Germany, Australia, France, Spain... The first step will be to prove the model and train the OCR to become better and better at it, including brand logo detection.
Skills: OCR Tesseract Data mining