数字图像处理课程项目：目标检测、CNNCBIR.zip

## 上传图片 ```rest method: post url: /api/start content-type: application/json body: { path string, preprocessMethod string, // "1-A" featureExtractionMethod string, //"2-A" similarityCalculationMethod string //"3-A" } response: { status: 200/error code } ``` - path是一个绝对路径，它是用户上传的本地图片路径 - 三个Method分别是预处理、特征提取、相似计算的算法名 - 我现在不知道这仨我们实现了啥算法叫啥名，需要算法大佬给个反馈 - 返回一个status，成功就200，失败就其他 检索处理完毕后，在 C:\Users\username\AppData\Local\Temp\SimpleCBIR\ 里输出一个result.json文件。 在写文件之前，在该目录放一个文件名为result.json.busy的文件，代表正在写文件。 写完后删除该.busy文件,代表解锁。 ## result.json ```rest { rawImage: string, // the raw path recallRatio: string, // "92.5%" 查全率 precision: string, // "87.5%" 查准率 position: string, // path of the target marked image features: { color: string, // path of the generated image texture: string, // path of the generated image shape: string // path of the generated image position: string // path of the target marked image }, results: { one: { similarity: string, //"99.9%" path: string //path of the most similar image, features: { color: string, // path of the generated image texture: string, // path of the generated image shape: string // path of the generated image } }, two: { similarity: string, //"76.2%" path: string, //path of the second similar image features: { color: string, // path of the generated image texture: string, // path of the generated image shape: string // path of the generated image } }, three: { similarity: string, //"64.1%" path: string, //path of the third similar image features: { color: string, // path of the generated image texture: string, // path of the generated image shape: string // path of the generated image } } } } ``` ## 获得图片 ```rest method: get url: /api/image query: ?i=path response: { status: 200/error code } ``` - path是result.json中获得的各个图片的path