mirror of
https://github.com/PaddlePaddle/FastDeploy.git
synced 2025-10-08 10:00:29 +08:00
English | 简体中文
PP-OCRv3 Frontend Deployment Example
This document introduces the deployment of PaddleOCR's PP-OCRv3 models to run in the browser, and the js interface in the @paddle-js-models/ocr npm package.
Frontend Deployment PP-OCRv3 Model
For PP-OCRv3 model web demo, refer to reference document
PP-OCRv3 js Interface
import * as ocr from "@paddle-js-models/ocr";
await ocr.init(detConfig, recConfig);
const res = await ocr.recognize(img, option, postConfig);
ocr model loading and initialization, where the model is in Paddle.js model format. For the conversion of js models, refer to the document
init function parameter
- detConfig(dict): The configuration parameter for the text detection model. Default {modelPath: 'https://js-models.bj.bcebos.com/PaddleOCR/PP-OCRv3/ch_PP-OCRv3_det_infer_js_960/model.json', fill: '#fff', mean: [0.485, 0.456, 0.406],std: [0.229, 0.224, 0.225]}; Among them, modelPath is the path of the text detection model, fill is the padding value in the image pre-processing, and mean/std are the mean and standard deviation in the pre-processing.
- recConfig(dict)): The configuration parameter for the text recognition model. Default {modelPath: 'https://js-models.bj.bcebos.com/PaddleOCR/PP-OCRv3/ch_PP-OCRv3_rec_infer_js/model.json', fill: '#000', mean: [0.5, 0.5, 0.5], std: [0.5, 0.5, 0.5]}; Among them, modelPath is the path of the text detection model, fill is the padding value in the image pre-processing, and mean/std are the mean and standard deviation in the pre-processing.
recognize function parameter
- img(HTMLImageElement): Enter an image parameter in HTMLImageElement.
- option(dict): The canvas parameter of the visual text detection box. No need to set.
- postConfig(dict): Text detection post-processing parameter. Default: {shape: 960, thresh: 0.3, box_thresh: 0.6, unclip_ratio:1.5}; thresh is the binarization threshold of the output prediction image. box_thresh is the threshold of the output box, below which the prediction box will be discarded. unclip_ratio is the expansion ratio of the output box.