Re: [tesseract-ocr] Beginner question : could not initialize tesseract, missing eng.traineddata file in tessdata

eng_pcb.traineddata is a traineddata starting with eng.traineddata i did lstm training to improve the detection of ocr rather than the recognition. i used tesstrain git repo. final error: couldn't find the legacy components in eng_pcb.traineddata On Monday, April 22, 2024 at 6:43:54 PM UTC+2

Re: [tesseract-ocr] Beginner question : could not initialize tesseract, missing eng.traineddata file in tessdata

If you used the tesstrain you trained the lstm engine. Why do you then ask tesseract to use a legacy engine? Do you understand what you are doing? Zdenko št 25. 4. 2024 o 11:35 Surya VaraPrasad Alla napísal(a): > eng_pcb.traineddata is a traineddata starting with eng.traineddata > > i did lstm

[tesseract-ocr] Challenge: need help extracting text from low quality image

Image: [image: Untitled.png] Current Code: #!/usr/bin/env node const path = require('path'); const sharp = require('sharp'); const { createWorker } = require('tesseract.js'); const [,, imagePath] = process.argv; const image = path.resolve('C:\\Users\\c\\attendance-system\\server\\ Untitled.png');