A Go library for solving hCaptchas with YOLOv3 and options for other image recognition systems within twenty seconds.
In order to use the main YOLOv3 solver, you'll need to install gocv, and the YOLOv3 config, weights, and names, which can be downloaded from the Go YOLOv3 repository, specifically the getModels.sh script, which will get all three for you.
Save all three files in a yolo directory in the main directory of your project. go-hcaptcha will use this directory
to load the YOLOv3 model.
In order to solve, you need the site URL (not the domain!), and the site key, which can be found in the HTML of the website with the hCaptcha challenge.
Below is a basic example of how to use the solver with the two using YOLOv3.
c, err := NewChallenge(siteUrl, siteKey)
if err != nil {
panic(err)
}
err = c.Solve(&YOLOSolver{Log: c.log})
if err != nil {
c.log.Panic(err)
}
c.log.Info(c.Token())The motion data capturing required with hCaptcha would not be possible without the work of @h0nde and his py-hcaptcha solver in Python.
There were quite a lot of changes with the hCaptcha API, so the solver was updated to reflect these changes, with the generous help of @aw1875 and his puppeteer-hcaptcha solver in JavaScript.
This project was inspired by the work of @JimmyLaurent and his hcaptcha-solver also in JavaScript. I'd like to thank him for his work, and for being a motivation to create this library.