DocumentsOCR

Screenshots

Requirements

• XCode 8 +
• iOS 8.0 +

Example

To run the example project, clone the repo, and run pod install from the Example directory first.

Used frameworks

DocumentOCR framework uses

• TesseractOCR text recognition of machine readable codes
• PodAsset easy way to use pod assets
• GPUImage helps to crop image from camera

Installation

DocumentsOCR is available through CocoaPods. To install it, simply add the following line to your Podfile:

pod "DocumentsOCR"

Usage

Import DocumentsOCR framework

    import DocumentsOCR

Implement DocumentScannerDelegate protocol

Create your custom class, which implemets DocumentScannerDelegate protocol

class ExampleViewController: UIViewController, DocumentScannerDelegate {
    	
    func documentScanner(scanner: DocumentScanner, willBeginScanningImage image: UIImage) {
        // do something with image
    }
    	
    func documentScanner(scanner: DocumentScanner, didFinishScanningWithInfo info: DocumentInfo) {
        // do something with DocumentInfo instance
    }	
    	
    func documentScanner(scanner: DocumentScanner, didFailWithError error: NSError) {
        // handle error
    }
    	
    // some other code here ...
}

Create DocumentScanner instance

Initialize DocumentScanner instance with references to UIViewController and DocumentScannerDelegate

	var scanner = DocumentScanner(containerVC: self, withDelegate: self)

Present camera controller

Scanner instance can present view controller with camera and border (actually, container view controller will do this inside document scanner instance).

	func someMethod() {
		scanner.presentCameraViewController()
	}

Events after "take shoot" button pressed

After take shoot button pressed, these delegate methods called:

    func documentScanner(scanner: DocumentScanner, willBeginScanningImage image: UIImage)

Then if image was recognized successfull:

    func documentScanner(scanner: DocumentScanner, didFinishScanningWithInfo info: DocumentInfo)

If some error happened

    func documentScanner(scanner: DocumentScanner, didFailWithError error: NSError)

Author

Michael Babaev, mmbabaev@gmail.com

License

DocumentsOCR is available under the MIT license. See the LICENSE file for more info.

Issues:

Before using DocumentOCR, you must set Enable bitcode value to "No" in tesseract framework (target TesseractOCRiOS -> Build settings -> Enable Bitcode)

TODO

• documentation
• fix minor UI defects in example
• code refactoring
• pod string for all versions (without using ~> version")
• improve mr code recognition
• check visa document recognitions
• unit tests for camera shoots
• take many pictures when "take shoot" button pressed, then choose best image for recognition