Skip to content

📄 Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.

clean the ocrweb folders

SWHLbfee26b609e8
diff --git a/ocrweb/main_redis.py b/ocrweb/main_redis.py
deleted file mode 100644
index e7d9df4..0000000
--- a/ocrweb/main_redis.py
+++ /dev/null
@@ -1,83 +0,0 @@
-# !/usr/bin/env python
-# -*- encoding: utf-8 -*-
-# @File: main.py
-# @Author: Max
-import base64
-import json
-import uuid
-from pathlib import Path
-
-import cv2
-import numpy as np
-import redis
-from flask import Flask, render_template, request
-from rq import Queue
-from rq.job import Job
-
-
-from task_redis import conn, detect_recognize
-
-
-root_path = Path(__file__).resolve().parent
-
-
-r = redis.Redis(host='localhost',
- port=6379,
- decode_responses=True)
-
-# 建立与Redis server的连接并初始化一个队列
-q = Queue(connection=conn, default_timeout=5000)
-
-app = Flask(__name__)
-
-# 设置上传文件大小
-app.config['MAX_CONTENT_LENGTH'] = 10 * 1024 * 1024 # 10M
-
-
-def index():
- return render_template('index_redis.html')
-
-
[email protected]('/ocr', methods=['POST', 'GET'])
-def ocr():
- if request.method == 'POST':
- # 获取图像数据
- url_get = request.get_json()
- url_get = str(url_get).split(',')[1]
- image = base64.b64decode(url_get)
- nparr = np.frombuffer(image, np.uint8)
- image = cv2.imdecode(nparr, cv2.IMREAD_COLOR)
-
- img_name = str(uuid.uuid4())
-
- image_directory = root_path / 'static' / 'images'
- if not image_directory.exists():
- image_directory.mkdir(parents=True, exist_ok=True)
-
- img_path = str(image_directory / f'{img_name}.jpg')
- cv2.imwrite(img_path, image)
-
- # 压入队列
- job = q.enqueue(detect_recognize,
- args=(img_path, ),
- result_ttl=5000)
- return json.dumps({'job_id': job.get_id()})
-
-
[email protected]('/get_results', methods=['POST'])
-def get_results():
- job_id = request.get_json()
- job = Job.fetch(job_id['job_id'], connection=conn)
- if job.is_finished:
- job_result = eval(job.result)
- job_result['status'] = True
- return json.dumps(job_result)
- else:
- return json.dumps({'status': False})
-
-
-if __name__ == '__main__':
- app.run(host='0.0.0.0', port=9003,
- debug=False, processes=True,
- threaded=True)
diff --git a/ocrweb/readme.md b/ocrweb/readme.md
index ec98dc5..1efc7df 100755
--- a/ocrweb/readme.md
+++ b/ocrweb/readme.md
@@ -3,7 +3,6 @@
```text
.
├── main.py
-├── main_redis.py
├── readme.md
├── resources
│ ├── bpocr.py
@@ -17,10 +16,9 @@
│ ├── css
│ ├── images
│ └── js
-├── task.py # 单独开启rq worker时使用
+├── task.py
└── templates
- ├── index.html
- └── index_redis.html # 带有redis队列web界面
+ └── index.html
```
#### Redis队列web界面使用步骤
diff --git a/ocrweb/requirements.txt b/ocrweb/requirements.txt
index 57a6d87..b40ec4c 100644
--- a/ocrweb/requirements.txt
+++ b/ocrweb/requirements.txt
@@ -1,11 +1,6 @@
-onnxruntime==1.5.2
-opencv-python==4.2.0.32
-numpy
-pyclipper
-Shapely
-imghdr
-python==3.6.4
-rq==1.7.0
-redis==3.5.3
-requests==2.25.1
-Flask==1.1.2
\ No newline at end of file
+onnxruntime==1.7.0
+opencv_python==4.5.1.48
+six==1.15.0
+pyclipper==1.2.1
+numpy==1.19.1
+Shapely==1.7.1
diff --git a/ocrweb/task.py b/ocrweb/task.py
index 80ae624..5449b9c 100644
--- a/ocrweb/task.py
+++ b/ocrweb/task.py
@@ -13,7 +13,7 @@ import numpy as np
from resources.bpocr import TextSystem, draw_text_det_res
# 实例化模型
-det_model_path = 'resources/models/ch_ppocr_server_v2.0_det_train.onnx'
+det_model_path = 'resources/models/ch_ppocr_server_v2.0_det_infer.onnx'
cls_model_path = 'resources/models/ch_ppocr_mobile_v2.0_cls_infer.onnx'
rec_model_path = 'resources/models/ch_ppocr_mobile_v2.0_rec_infer.onnx'
diff --git a/ocrweb/task_redis.py b/ocrweb/task_redis.py
deleted file mode 100644
index d34c596..0000000
--- a/ocrweb/task_redis.py
+++ /dev/null
@@ -1,71 +0,0 @@
-# !/usr/bin/env python
-# -*- encoding: utf-8 -*-
-# @File: worker.py
-# @Time: 2021/03/07 20:29:32
-# @Author: Max
-import json
-import base64
-import cv2
-import time
-import numpy as np
-
-import redis
-from rq import Worker, Queue, Connection
-from resources.bpocr import TextSystem, draw_text_det_res
-
-listen = ['default']
-redis_url = "redis://localhost:6379" # redis server 默认地址
-conn = redis.from_url(redis_url)
-
-# 实例化模型
-det_model_path = 'resources/models/ch_ppocr_server_v2.0_det_infer.onnx'
-cls_model_path = 'resources/models/ch_ppocr_mobile_v2.0_cls_infer.onnx'
-rec_model_path = 'resources/models/ch_ppocr_mobile_v2.0_rec_infer.onnx'
-
-text_sys = TextSystem(det_model_path,
- rec_model_path,
- use_angle_cls=True,
- cls_model_path=cls_model_path)
-
-
-def detect_recognize(image_path):
- if isinstance(image_path, str):
- image = cv2.imread(image_path)
- elif isinstance(image_path, np.ndarray):
- image = image_path
-
- t1 = time.time()
- dt_boxes, rec_res, img = text_sys(image)
- t2 = time.time()
-
- if dt_boxes is None or rec_res is None:
- temp_rec_res = []
- rec_res_data = json.dumps(temp_rec_res,
- indent=2,
- ensure_ascii=False)
- elapse = 0
- image = cv2.imencode('.jpg', img)[1]
- img = str(base64.b64encode(image))[2:-1]
- else:
- temp_rec_res = []
- for i, value in enumerate(rec_res):
- temp_rec_res.append([i, value[0], value[1]])
- temp_rec_res = np.array(temp_rec_res)
- rec_res_data = json.dumps(temp_rec_res.tolist(),
- indent=2,
- ensure_ascii=False)
-
- det_im = draw_text_det_res(dt_boxes, img)
- image = cv2.imencode('.jpg', det_im)[1]
- img = str(base64.b64encode(image))[2:-1]
-
- elapse = f'{t2-t1:.3f}'
- return json.dumps({'image': img,
- 'elapse': elapse,
- 'rec_res': rec_res_data})
-
-
-if __name__ == '__main__':
- with Connection(conn): # 建立与redis server的连接
- worker = Worker(list(map(Queue, listen))) # 建立worker监听给定的队列
- worker.work()
\ No newline at end of file
diff --git a/ocrweb/templates/index_redis.html b/ocrweb/templates/index_redis.html
deleted file mode 100644
index 8be5506..0000000
--- a/ocrweb/templates/index_redis.html
+++ /dev/null
@@ -1,178 +0,0 @@
-<!DOCTYPE html>
-<html lang="en">
-<head>
- <meta charset="UTF-8">
- <title>RapidOCR Demo</title>
- <link rel=stylesheet type="text/css" href="{{url_for('static', filename='css/main.css')}}">
-</head>
-<body>
- <h1 align="center">
- RapidOCR Demo &nbsp;
- <a href="https://github.com/znsoftm/RapidOCR" style="font-size: 14px;">欢迎Star</a>
- </h1>
- <div class="area">
- <span class="uplodNote">图片文件类型支持PNG、JPG、JPEG、BMP,图片大小不超过10M。</span>
- <span class="btn-gen" onclick="javascript:$('#upload_img').click();">点击上传</span>
- <input type="file" style="display: none;" name="pic" id="upload_img"/>
- <span id="status_upload" class="uplodNote"></span>
- &nbsp;
- <span class="btn-gen" onclick="javascript:$('#detect_recognize').click();" id="span_det_rec">开始识别</span>
- <input type="button" style="display: none;" name="detect" id="detect_recognize"/>
- </div>
- <br/>
- <div class="area">
- <div class="leftarea" id="detectDiv">
- <span class="span_title">文本检测结果:</span><br/><br/>
- <div class="wrapper" id="wrapper" hidden></div>
- <img id="detect_img" src=""/>
- </div>
-
- <div class="rightarea" id="recogDiv">
- <span id="rec_res" class="span_title">文本识别结果:</span>
- <br/> <br/>
- <table id="locTable" border="1" class="table" cellspacing="0" cellpadding="5"></table>
- </div>
- </div>
- <script src="{{url_for('static', filename='js/jquery-3.0.0.min.js')}}"></script>
- <script type="text/javascript">
- // 页面加载时,执行
- window.onload = function () {
- $("#detect_img").attr("width", "");
- $("#detect_img").attr("height", "");
- $("#span_det_rec").hide();
- }
- let job_id;
-
- $("#upload_img").change(function(){
- let file = document.getElementById("upload_img").files[0];
-
- // 判断图像格式是否匹配
- let imageName = file.name;
- let index = imageName.lastIndexOf(".");
- let extName = imageName.substr(index + 1);
- let imgArr = ["jpg", "bmp", "png", "jpeg"];
- if (!(imgArr.includes(extName.toLocaleLowerCase()))){
- alert("图像文件格式不支持!");
- return;
- }
-
- // 判断图像大小是否符合
- let imageSize = file.size / 1024 / 1024;
- if (imageSize >= 10) {
- alert("图像大小超过10M!");
- return;
- }
-
- var reader = new FileReader();
- reader.onload = function(e){
- var upload_data = {"file": reader.result};
- $.ajax(
- {
- url:"/ocr",
- type:"POST",
- data: JSON.stringify(upload_data),
- contentType: "application/json; charset=UTF-8",
- dataType: "json",
-
- beforeSend:function(){
- $("#detect_img").attr("src", "");
- $("#detect_img").hide();
-
- $("#status_upload").show();
- $("#status_upload").html("上传中...");
-
- $("#rec_res").html("文本识别结果:");
- $("#locTable").hide();
- },
- success:function (data) {
- $("#detect_img").show();
- $("#detect_img").attr("src", reader.result);
- $("#detect_img").attr("width", "95%");
- $("#detect_img").attr("height", "100%");
-
- $("#status_upload").hide();
- $("#span_det_rec").show();
-
- job_id = data["job_id"];
- }
- }
- );
- }
- reader.readAsDataURL(this.files[0])
- });
-
- $("#detect_recognize").click(function() {
- get_result();
- })
-
- function get_result() {
- $.ajax({
- type: "POST",
- url: "/get_results",
- data: JSON.stringify({"job_id": job_id}),
- contentType: "application/json; charset=UTF-8",
- dataType: "json",
- beforeSend:function(){
- $("#wrapper").show();
- },
- success: function (data) {
- if (data.status){
- $("#wrapper").hide();
- $("#span_det_rec").hide();
-
- if (data["image"]){
- set_image(data["image"]);
- }
- if (data["elapse"]){
- set_elapse(data["elapse"]);
- }
- if (data["rec_res"]){
- set_result(data["rec_res"]);
- $("#locTable").show();
- }
- }else{
- $("#wrapper").show();
- get_result();
- }
- }
- })
- }
-
- function set_image(image) {
- $("#detect_img").attr("src", "data:image/jpeg;base64,"+ image);
- $("#detect_img").show();
- }
-
- function set_elapse(elapse) {
- document.getElementById("rec_res").textContent = "文本识别结果(耗时:" + String(elapse) + "s):"
- }
-
- function set_result(rec_res) {
- var rec_res = JSON.parse(rec_res);
- // 绘制整个结果表格
- var data = "";
- data += "<tr>" +
- "<th bgcolor='#C0C0C0'>序号</th>" +
- "<th bgcolor='#C0C0C0'>识别结果</th>" +
- "<th bgcolor='#C0C0C0'>置信度</th>" +
- "</tr>"
- for (let i = 0; i < rec_res.length; i++) {
- const element = rec_res[i];
- let num = element[0];
-
- let rec_result = element[1];
-
- let score = Number(element[2]);
- score = score.toFixed(4);
-
- data += "<tr>"
- data += "<td>" + num + "</td>";
- data += "<td align='left'>" + rec_result + "</td>";
- data += "<td>" + score + "</td>";
- data += "</tr>"
- }
- document.getElementById("locTable").innerHTML = data;
- }
- </script>
-</body>
-</html>
\ No newline at end of file