Python Flask搭建yolov3目标检测系统详解流程

【人工智能项目】Python Flask搭建yolov3目标检测系统

后端代码

from flask import Flask, request, jsonify
from PIL import Image
import numpy as np
import base64
import io
import os

from backend.tf_inference import load_model, inference

os.environ['CUDA_VISIBLE_DEVICES'] = '0'

sess, detection_graph = load_model()

app = Flask(__name__)

@app.route('/api/', methods=["POST"])
def main_interface():
   response = request.get_json()
   data_str = response['image']
   point = data_str.find(',')
   base64_str = data_str[point:]  # remove unused part like this: "data:image/jpeg;base64,"

image = base64.b64decode(base64_str)      
   img = Image.open(io.BytesIO(image))

if(img.mode!='RGB'):
       img = img.convert("RGB")

# convert to numpy array.
   img_arr = np.array(img)

# do object detection in inference function.
   results = inference(sess, detection_graph, img_arr, conf_thresh=0.7)
   print(results)

return jsonify(results)

@app.after_request
def add_headers(response):
   response.headers.add('Access-Control-Allow-Origin', '*')
   response.headers.add('Access-Control-Allow-Headers', 'Content-Type,Authorization')
   return response

if __name__ == '__main__':
   app.run(debug=True, host='0.0.0.0')

展示部分

python -m http.server

python app.py

前端展示部分

来源：https://blog.csdn.net/Mind_programmonkey/article/details/121051114