https://github.com/tucan9389/textdetection-coreml

https://github.com/tucan9389/textdetection-coreml

Last synced: 4 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/tucan9389/textdetection-coreml
• Owner: tucan9389
• License: mit
• Created: 2019-02-20T16:10:18.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2019-02-20T20:05:53.000Z (over 6 years ago)
• Last Synced: 2025-04-30T07:47:05.619Z (5 months ago)
• Language: Swift
• Size: 18.7 MB
• Stars: 81
• Watchers: 4
• Forks: 9
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# TextDetection-CoreML

![platform-ios](https://img.shields.io/badge/platform-ios-lightgrey.svg)

![swift-version](https://img.shields.io/badge/swift-4.2-red.svg)

![lisence](https://img.shields.io/badge/license-MIT-black.svg)

This project is Text Detection on iOS using [Vision](https://developer.apple.com/documentation/vision) built-in model.
If you are interested in iOS + Machine Learning, visit [here](https://github.com/motlabs/iOS-Proejcts-with-ML-Models) you can see various DEMOs.


![TextDetection-CoreML_DEMO001](resource/TextDetection-CoreML_DEMO001.gif)

## Requirements

- Xcode 9.2+

- iOS 12.0+

- Swift 4.2

## Performance

### Inference Time

| device   | inference time |

| -------- | -------------- |

| iPhone X | 10 ms          |

## Build & Run

### 1. Prerequisites

#### Add permission in info.plist for device's camera access

![prerequest_001_plist](resource/prerequest_001_plist.png)

### 2. Dependencies

No external library yet.

### 3. Code

#### 3.1 Import Vision framework

```swift

import Vision

```

#### 3.2 Define properties for Vision

```swift

// properties on ViewController

var request: VNDetectTextRectanglesRequest?

```

#### 3.3 Configure and prepare

```swift

override func viewDidLoad() {

    super.viewDidLoad()

	let request = VNDetectTextRectanglesRequest(completionHandler: self.visionRequestDidComplete)

    request.reportCharacterBoxes = true

    self.request = request

}

func visionRequestDidComplete(request: VNRequest, error: Error?) { 

    /* ------------------------------------------------------ */

    /* something postprocessing what you want after inference */

    /* ------------------------------------------------------ */

}

```

#### 3.4 Inference 🏃‍♂️

```swift

// on the inference point

let handler = VNImageRequestHandler(cvPixelBuffer: pixelBuffer)

if let request = request {

	try? handler.perform([self.request])

}

```