基于YOLOv8和FASTAPI的目标检测后端API

# yolov8-fastapi This is an API for detecting LPG inspection images using YOLOv8 and FastAPI. ## Quickstart Before using this repository! Please replace the weight file /data/model/best.pt with your own trained model, unless you want to detect LPG inspection images. ## Installation Build and start YOLOv8 with FASTAPI on http://localhost:9099 ### Build ```docker-compose up``` ### Rebuild ```docker-compose up --build``` ## Directories ``` --/app |--/data |--/download # downloaded images |--/predict # predicted images and json files |--/model # weight file ``` ## Usage ### Models There are 2 Models in model.py: ``` # this is for user to auth class User(BaseModel): username: str password: str # this is for image to predict class ImageToPredict(BaseModel): type: str = 'url' # or 'base64' url: str = None base64: str = None date: str = None ``` ### API #### Login ``` http://localhost:9099/token ``` #### Predict ``` http://localhost:9099/token ``` Accepts URL or base64 image, and returns the prediction result in JSON format. ## Response ``` { "original_img": , "predict_img": , "result": result[0].tojson() } ``` ### Users Users are stored in table ```users``` of ```./sysuser.sqlite```, and the default username and password are ```admin``` and ```admin```. 资源来源于网络分享，如有侵权，请告知。仅供学习交流使用，禁止用于商业用途。