https://github.com/danlock/gogosseract
A reimplementation of https://github.com/otiai10/gosseract without CGo, running Tesseract compiled to WASM with Wazero
https://github.com/danlock/gogosseract
Last synced: 10 months ago
JSON representation
A reimplementation of https://github.com/otiai10/gosseract without CGo, running Tesseract compiled to WASM with Wazero
- Host: GitHub
- URL: https://github.com/danlock/gogosseract
- Owner: Danlock
- License: apache-2.0
- Created: 2023-10-28T02:19:06.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-11-07T06:05:17.000Z (over 2 years ago)
- Last Synced: 2024-11-14T11:39:57.175Z (over 1 year ago)
- Language: Go
- Size: 25.2 MB
- Stars: 141
- Watchers: 1
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# gogosseract
[](https://goreportcard.com/report/github.com/danlock/gogosseract)
[](https://pkg.go.dev/github.com/danlock/gogosseract)
A reimplementation of https://github.com/otiai10/gosseract without CGo, running Tesseract compiled to WASM with Emscripten via Wazero.
Tesseract is an Optical Character Recognition library written in C++.
The WASM is generated from my [personal](https://github.com/Danlock/tesseract-wasm) fork of robertknight's well written tesseract-wasm project.
Note that Tesseract is only compiled with support for the LSTM neural network OCR engine, and not for "classic" Tesseract.
> [!CAUTION]
> This library and it's dependent libraries was broken by a backwards incompatible change in wazero 1.8.0. This library will not be updated. If you plan on
> using this library regardless, make sure your dependencies are the same version as what's in the go.mod file in this repo.
> Also, CGO gosseract is like 6 times faster than this library anyway the last time I checked.
# Training Data
Tesseract requires training data in order to accurately recognize text. The official source is [here](https://github.com/tesseract-ocr/tessdata_fast). Strategies for dealing with this include downloading it at runtime, or embedding the file within your Go binary using go:embed at compile time.
# Accuracy
Tesseract can work better if the input images are preprocessed. See this page for tips.
https://tesseract-ocr.github.io/tessdoc/ImproveQuality.html
# Examples
Using Tesseract to parse text from an image.
```go
trainingDataFile, err := os.Open("eng.traineddata")
handleErr(err)
cfg := gogosseract.Config{
Language: "eng",
TrainingData: trainingDataFile,
}
// While Tesseract's logs are very useful for debugging, you have the option to silence or redirect it
cfg.Stderr = io.Discard
cfg.Stdout = io.Discard
// Compile the Tesseract WASM and run it, loading in the TrainingData and setting any Config Variables provided
tess, err := gogosseract.New(ctx, cfg)
handleErr(err)
imageFile, err := os.Open("image.png")
handleErr(err)
err = tess.LoadImage(ctx, imageFile, gogosseract.LoadImageOptions{})
handleErr(err)
text, err = tess.GetText(ctx, func(progress int32) { log.Printf("Tesseract parsing is %d%% complete.", progress) })
handleErr(err)
// Closing the Tesseract instance will clean up everything used by Tesseract and it's WASM module
handleErr(tess.Close(ctx))
```
Using a Pool of Tesseract workers for thread safe concurrent image parsing.
```go
cfg := gogosseract.Config{
Language: "eng",
TrainingData: trainingDataFile,
}
// Create 10 Tesseract instances that can process image requests concurrently.
pool, err := gogosseract.NewPool(ctx, 10, gogosseract.PoolConfig{Config: cfg})
handleErr(err)
// ParseImage loads the image and waits until the Tesseract worker sends back your result.
hocr, err := pool.ParseImage(ctx, img, gogosseract.ParseImageOptions{
IsHOCR: true,
})
handleErr(err)
// Always remember to Close the pool to release resources
handleErr(pool.Close())
```