FORMAT_CONVERSION_MAXCOMPUTE/ocr/ocr_interface.py

import base64
import io
import json
import multiprocessing as mp
import socket
import sys
import os

from PIL import Image

sys.path.append(os.path.dirname(os.path.abspath(__file__)) + "/../")
import time
import traceback
from multiprocessing.context import Process
import cv2
import requests
import logging
import numpy as np
os.environ['FLAGS_eager_delete_tensor_gb'] = '0'
from format_convert.utils import request_post, test_gpu, get_intranet_ip, log, get_md5_from_bytes, bytes2np
from flask import Flask, request
from format_convert import _global


# 接口配置
app = Flask(__name__)


use_angle_cls = False


@app.route('/ocr', methods=['POST'])
def _ocr():
    _global._init()
    _global.update({"port": globals().get("port")})
    start_time = time.time()

    log("into ocr_interface _ocr")
    try:
        if not request.form:
            log("ocr no data!")
            return json.dumps({"text": str([-9]), "bbox": str([-9])})
        data = request.form.get("data")
        _md5 = request.form.get("md5")
        only_rec = request.form.get("only_rec")
        if only_rec is None:
            only_rec = 0
        else:
            only_rec = int(only_rec)
        _global.update({"md5": _md5})
        ocr_model = globals().get("global_ocr_model")
        if ocr_model is None:
            log("----------- init ocr_model ------------")
            ocr_model = OcrModels().get_model()
            globals().update({"global_ocr_model": ocr_model})
        text = ocr(data, ocr_model, only_rec)
        return json.dumps(text)
    except TimeoutError:
        return json.dumps({"text": str([-5]), "bbox": str([-5])})
    except:
        traceback.print_exc()
        return json.dumps({"text": str([-1]), "bbox": str([-1])})
    finally:
        log("ocr interface finish time " + str(time.time()-start_time))


def ocr(data, ocr_model, only_rec=0):
    log("into ocr_interface ocr")
    try:
        img_data = base64.b64decode(data)
        text = picture2text(img_data, ocr_model, only_rec)
        return text
    except TimeoutError:
        return {"text": str([-5]), "bbox": str([-5])}


def picture2text(img_data, ocr_model, only_rec=0):
    log("into ocr_interface picture2text")
    try:
        # 二进制数据流转np.ndarray [np.uint8: 8位像素]
        img = bytes2np(img_data)
        # cv2.imwrite('ocr.jpg', img)

        # 预测
        if only_rec:
            results = ocr_model.ocr(img, det=False, rec=True, cls=use_angle_cls)
        else:
            results = ocr_model.ocr(img, det=True, rec=True, cls=use_angle_cls)

        # 循环每张图片识别结果
        text_list = []
        bbox_list = []
        if only_rec:
            text_list = [results[0][0]]
            bbox_list = []
        else:
            for line in results:
                text_list.append(line[-1][0])
                bbox_list.append(line[0])
        return {"text": str(text_list), "bbox": str(bbox_list)}

    except TimeoutError:
        raise TimeoutError
    except Exception:
        log("picture2text error!")
        traceback.print_exc()
        return {"text": str([]), "bbox": str([])}


def get_best_predict_size(image_np):
    sizes = [1280, 1152, 1024, 896, 768, 640, 512, 384, 256, 128]

    min_len = 10000
    best_height = sizes[0]
    for height in sizes:
        if abs(image_np.shape[0] - height) < min_len:
            min_len = abs(image_np.shape[0] - height)
            best_height = height

    min_len = 10000
    best_width = sizes[0]
    for width in sizes:
        if abs(image_np.shape[1] - width) < min_len:
            min_len = abs(image_np.shape[1] - width)
            best_width = width

    return best_height, best_width


class OcrModels:
    def __init__(self):
        from ocr.paddleocr import PaddleOCR
        try:
            log('----------- init ocr model ---------------')
            self.ocr_model = PaddleOCR(use_angle_cls=use_angle_cls, lang="ch")
        except:
            print(traceback.print_exc())
            raise RuntimeError

    def get_model(self):
        return self.ocr_model


def test_ocr_model(from_remote=True):
    file_path = "error8.png"
    file_path = "C:/Users/Administrator/Downloads/dbf46fe38862ac03209f1b2c12b1adc1.jpg"
    with open(file_path, "rb") as f:
        file_bytes = f.read()

    src = """
    data:image/png;base64,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    
    """
    image_data = src.split('data:image/png;base64,')[1]
    # 解码 base64 字符串
    file_bytes = base64.b64decode(image_data)

    file_base64 = base64.b64encode(file_bytes)
    _md5 = get_md5_from_bytes(file_bytes)[0]
    only_rec = 0
    _global._init()
    _global.update({"port": 15010, "md5": _md5})

    if from_remote:
        file_json = {"data": file_base64, "md5": _md5, 'only_rec': only_rec}
        # _url = "http://192.168.2.102:17000/ocr"
        # _url = "http://127.0.0.1:17000/ocr"
        _url = "http://120.132.118.205:17000/ocr"
        print(json.loads(request_post(_url, file_json)))
    else:
        ocr_model = OcrModels().get_model()
        result = ocr(file_base64, ocr_model, only_rec=only_rec)
        text = result.get('text')
        bbox = result.get('bbox')
        print('bbox', bbox)
        print(result)


if __name__ == '__main__':
    test_ocr_model()

#     src = """
# data:image/png;base64,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
# """
#
#     image_data = src.split('data:image/png;base64,')[1]
#
#     # 解码 base64 字符串
#     image_bytes = base64.b64decode(image_data)
#
#     # 将字节转换为图像
#     # image = Image.open(io.BytesIO(image_bytes))
#
#     # image.show('img')
#
#     # with open(r'C:\Users\Administrator\Desktop\test_image\error16.jpg', 'rb') as f:
#     #     image_bytes = f.read()
#
#     image = bytes2np(image_bytes)
#
#     cv2.imshow('img', image)
#     cv2.imwrite('./1.png', image)
#     cv2.waitKey(0)