📄 Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.
Add ocrweb api deploy
SWHL96efec9a0a7e
diff --git a/.gitignore b/.gitignoreindex 5aefa29..80bb95f 100755--- a/.gitignore+++ b/.gitignore@@ -151,4 +151,6 @@ long1.jpg*.pdiparams*.pdiparams.info-*.pdmodel\ No newline at end of file+*.pdmodel++.DS_Store\ No newline at end of filediff --git a/README.md b/README.mdindex f2671ec..0b4d198 100755--- a/README.md+++ b/README.md@@ -25,6 +25,7 @@- [RapidOCR (捷智OCR)](#rapidocr-捷智ocr)- [简介](#简介)- [近期更新(more)](#近期更新more)+ - [🦣2022-08-14 update:](#2022-08-14-update)- [🎧2022-07-10 update:](#2022-07-10-update)- [✨2022-07-07 update:](#2022-07-07-update)- [相关项目(more)](#相关项目more)@@ -62,6 +63,10 @@- 基于百度的开源PaddleOCR 模型及训练,任何人可以使用本推理库,也可以根据自己的需求使用百度的paddlepaddle框架进行模型优化。## 近期更新([more](./docs/change_log.md))+#### 🦣2022-08-14 update:+- ocrweb部分增加以API方式部署调用的功能,可以通过发送POST请求,来获得OCR识别结果。+- 详情参见:[API方式调用](./ocrweb/README.md#以api方式运行和调用)+#### 🎧2022-07-10 update:- 添加onnxruntime-gpu的测试情况→[link](./python/README.md#onnxruntime-gpu版推理配置)- 添加benchamark测试集 → [link](./images/README.md)diff --git a/docs/README_en.md b/docs/README_en.mdindex 41d5c9f..f177e7f 100644--- a/docs/README_en.md+++ b/docs/README_en.md@@ -24,6 +24,7 @@- [RapidOCR](#rapidocr)- [Introduction](#introduction)- [Recently updates(more)](#recently-updatesmore)+ - [🦣2022-08-14 update:](#2022-08-14-update)- [🎧2022-07-10 update:](#2022-07-10-update)- [✨2022-07-07 update:](#2022-07-07-update)- [Related Projects(more)](#related-projectsmore)@@ -61,6 +62,10 @@- Based on Baidu's open source PaddleOCR model and training, anyone can use this inference library, or use Baidu's paddlepaddle framework for model optimization according to their own needs.## Recently updates([more](./change_log_en.md))+#### 🦣2022-08-14 update:+- The ocrweb part adds the function of deploying calls by API, and you can send POST requests to get OCR recognition results.+- For details, see: [API deploy](./ocrweb/README.md#以api方式运行和调用)+#### 🎧2022-07-10 update:- Add test case for onnxruntime-gpu → [link](../python/README.md#onnxruntime-gpu version inference configuration)- Add benchamark test set → [link](../images/README.md)diff --git a/docs/change_log.md b/docs/change_log.mdindex 94d005b..e727e4a 100644--- a/docs/change_log.md+++ b/docs/change_log.md@@ -1,3 +1,7 @@+#### 🦣2022-08-14 update:+- ocrweb部分增加以API方式部署调用的功能,可以通过发送POST请求,来获得OCR识别结果。+- 详情参见:[API方式调用](../ocrweb/README.md#以api方式运行和调用)+#### ✨2022-07-07 update:- 修复python版中v3 rec推理bug,并将v3 rec与v2 rec合并为同一套推理代码,更加简洁和方便- 添加python模块下的单元测试diff --git a/docs/change_log_en.md b/docs/change_log_en.mdindex 19ac040..c301a52 100644--- a/docs/change_log_en.md+++ b/docs/change_log_en.md@@ -1,3 +1,7 @@+#### 🦣2022-08-14 update:+- The ocrweb part adds the function of deploying calls by API, and you can send POST requests to get OCR recognition results.+- For details, see: [API deploy](../ocrweb/README.md#以api方式运行和调用)+#### 🎧2022-07-10 update:- Add test case for onnxruntime-gpu → [link](../python/README.md#onnxruntime-gpu version inference configuration)- Add benchamark test set → [link](../images/README.md)diff --git a/ocrweb/README.md b/ocrweb/README.mdindex bbdedfd..31c3e5f 100755--- a/ocrweb/README.md+++ b/ocrweb/README.md@@ -10,7 +10,7 @@<img src="../assets/ocrweb_time.jpg" width="80%" height="80%"></div>-### 运行+### Web方式运行1. 安装`requirements.txt`下相关包```shellpip install -r requirements.txt -i https://pypi.douban.com/simple/@@ -50,3 +50,58 @@python main.py```4. 打开`http://0.0.0.0:9003/`即可, enjoy it!++### 以API方式运行和调用+1. 同**Web方式运行**中步骤1+2. 同**Web方式运行**中步骤2+3. 运行`api.py`+ ```python+ python api.py+ ```++4. 发送post请求,调用+ ```python+ import ast+ import base64+ import json++ import requests+++ def get_byte(img_path):+ with open(img_path, 'rb') as f:+ img_byte = base64.b64encode(f.read())+ img_str = img_byte.decode('ascii')+ return img_str+++ if __name__ == '__main__':+ header = {+ "cookie": "token=code_space;",+ "Accept": "*/*",+ "Accept-Encoding": "gzip, deflate, br",+ "Accept-Language": "zh-CN,zh;q=0.9",+ "Connection": "keep-alive",+ "Content-Type": "application/json; charset=UTF-8",+ }++ url = 'http://localhost:9003/ocr'++ img_path = '../images/1.jpg'+ img = get_byte(img_path)+ post_json = json.dumps({'file': img})++ response = requests.post(url, data=post_json, headers=header)++ if response.status_code == 200:+ rec_res = ast.literal_eval(response.text)+ print(rec_res)+ else:+ print(response.status_code)++ ```++5. 输出以下结果,即为正确。+ ```text+ [['0', '香港深圳抽血', '0.93583983'], ['1', '专业查性别', '0.89865875'], ['2', '专业鉴定B超单', '0.9955703'], ['3', 'b超仪器查性别', '0.99489486'], ['4', '加微信eee', '0.99073666'], ['5', '可邮寄', '0.99923944']]+ ```\ No newline at end of filediff --git a/ocrweb/api.py b/ocrweb/api.pynew file mode 100644index 0000000..21dd6b6--- /dev/null+++ b/ocrweb/api.py@@ -0,0 +1,44 @@+# -*- encoding: utf-8 -*-+# @Author: SWHL+# @Contact: [email protected]+import base64+import json+from wsgiref.simple_server import make_server++import cv2+import numpy as np+from flask import Flask, render_template, request++from task import detect_recognize++app = Flask(__name__)+app.config['MAX_CONTENT_LENGTH'] = 3 * 1024 * 1024++[email protected]('/')+def index():+ return render_template('index.html')++[email protected]('/ocr', methods=['POST'])+def ocr():+ if request.method == 'POST':+ url_get = request.get_json()+ img_str = url_get.get('file')++ image = base64.b64decode(img_str)+ nparr = np.frombuffer(image, np.uint8)+ image = cv2.imdecode(nparr, cv2.IMREAD_COLOR)+ if image.ndim == 2:+ image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR)++ img, elapse, elapse_part, rec_res_data = detect_recognize(image)+ return rec_res_data+++if __name__ == '__main__':+ ip = '0.0.0.0'+ ip_port = 9003++ server = make_server(ip, ip_port, app)+ server.serve_forever()diff --git a/ocrweb/main.py b/ocrweb/main.pyindex 928208b..380592f 100644--- a/ocrweb/main.py+++ b/ocrweb/main.py@@ -2,16 +2,16 @@# @Author: SWHL# @Contact: [email protected]import base64+import json+from wsgiref.simple_server import make_serverimport cv2import numpy as npfrom flask import Flask, render_template, request-from wsgiref.simple_server import make_serverfrom task import detect_recognizeapp = Flask(__name__)-app.config['MAX_CONTENT_LENGTH'] = 3 * 1024 * 1024@@ -24,14 +24,19 @@ def index():def ocr():if request.method == 'POST':url_get = request.get_json()- url_get = str(url_get).split(',')[1]+ img_str = str(url_get).split(',')[1]- image = base64.b64decode(url_get)+ image = base64.b64decode(img_str)nparr = np.frombuffer(image, np.uint8)image = cv2.imdecode(nparr, cv2.IMREAD_COLOR)if image.ndim == 2:image = cv2.cvtColor(image, cv2.COLOR_GRAY2BGR)- return detect_recognize(image)++ img, elapse, elapse_part, rec_res_data = detect_recognize(image)+ return json.dumps({'image': img,+ 'total_elapse': f'{elapse:.4f}',+ 'elapse_part': elapse_part,+ 'rec_res': rec_res_data})if __name__ == '__main__':diff --git a/ocrweb/task.py b/ocrweb/task.pyindex 0d813e5..6b3c989 100644--- a/ocrweb/task.py+++ b/ocrweb/task.py@@ -50,10 +50,7 @@ def detect_recognize(image_path):elapse = reduce(lambda x, y: float(x)+float(y), elapse_part)elapse_part = ','.join([str(x) for x in elapse_part])- return json.dumps({'image': img,- 'total_elapse': f'{elapse:.4f}',- 'elapse_part': elapse_part,- 'rec_res': rec_res_data})+ return img, elapse, elapse_part, rec_res_datadef check_and_read_gif(img_path):