# IdentifyGraph

**Repository Path**: BHVHGHJ/IdentifyGraph

## Basic Information

- **Project Name**: IdentifyGraph
- **Description**: No description available
- **Primary Language**: Android
- **License**: Apache-2.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2018-07-08
- **Last Updated**: 2021-07-23

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

## 基于Android平台的TensorFlew图像识别相机

[TOC]

[TensorFlew:](https://blog.csdn.net/masa_fish/article/details/54097796/)深度学习模型支持三个领域的模型训练

1. TF Classify: 实时物体分类
2. TF Detect: 实时行人检测
3. TF Stylize: 实时图片风格迁移

本产品实现的是**TF Classify**到Android平台:iphone:移植。

### 需求分析

* TensorFlew 模型集成
* JNI Java平台与C++协调
* Camera :camera: 相关功能集成



### 可行性分析--略:smile:

### 概要设计

​	采用TensorFlew已经训练好的分类器:muscle:，获取当前相机:camera:拍摄到的图片:framed_picture:，进行分类识别，最后将结果集返回到UI界面显示，其中结果反应的是当前图片中最有可能是某三种（以内）物体的概论。

**流程图**

```flow
start=>start: Start

buttonOnclick=>inputoutput: Capture Picture Button Onclicklistener trigger

camera capture=>subroutine: Camera capture Picture

tensorFlew analysis=>subroutine: TensorFlew analysis picture

imageView=>inputoutput: show the captured picture from Camera Kit

textView=>inputoutput: show the resultSet from tensorFlew

end=>end: End

start->buttonOnclick->camera capture->tensorFlew analysis->imageView->textView->end


```



**时序图**

  ```sequence
participant UI
participant Model
participant Camera Kit
participant TensorFlew

UI->Camera Kit:Capture()

Model->TensorFlew: feed(File model,File label)

Camera Kit->TensorFlew: feed(Bitmao picture)

Camera Kit -->UI:show(Bitemap picture)

TensorFlew-->UI:handMessage(String[] result)
  ```



TensorFlew feed

```java
 /**
     * recognize image
     * @param bitmap  The asset of picture to recognize
     */

public List<Recognition> recognizeImage(final Bitmap bitmap) {
    
}
```



```java
 /**
     * Initializes a native TensorFlow session for classifying images.
     *
     * @param assetManager  The asset manager to be used to load assets.
     * @param modelFilename The filepath of the model GraphDef protocol buffer.
     * @param labelFilename The filepath of label file for classes.
     * @param inputSize     The input size. A square image of inputSize x inputSize is assumed.
     * @param imageMean     The assumed mean of the image values.
     * @param imageStd      The assumed std of the image values.
     * @param inputName     The label of the image input node.
     * @param outputName    The label of the output node.
     * @throws IOException
     */

public static Classifier create(
            AssetManager assetManager,
            String modelFilename,
            String labelFilename,
            int inputSize,
            int imageMean,
            float imageStd,
            String inputName,
            String outputName){
    
}
```



**调用CameraKit库方法**

```java
 @Override
public void start(){
    
}

@Override
public void stop(){
    
}

 @Override
public void captureImage(new CameraKitEventCallback<CameraKitImage>() {

}
                        
 @Override
public void callback(CameraKitImage cameraKitImage) {
                    
}
```

### 详细设计

**界面**

- 图标

![72C54EAA76CA60579666589D2CC57910](./72C54EAA76CA60579666589D2CC57910.jpg)

- 主界面展示，其中：
  - camera 区域为相机当前面向的物体
  - image 区域为用户想识别的物体照片
  - result 区域为照片的识别结果，每一条表示可能的物体以及其概论
  - 左下角图标：前后摄像头选择
  - 中下角图标：获取对应分类物体的照片按钮，同时展示分类效果

![主界面](./主界面.png)

![DE547807C0008910B7FCC72BF1772497](./DE547807C0008910B7FCC72BF1772497.jpg)

### 产品使用说明

  正常相机怎么用就怎么用:sweat_smile: