花一个小时，用Flask写一个简单的行人检测API
花一个小时，用Flask写一个简单的行人检测接口
- 前提条件
- 实验环境
- 项目结构
- 主要代码
- 运行结果
- 参考链接
前提条件
- 了解Python语言，并会安装第三方库
- 了解Python Web Flask框架
- 了解PyTorch深度学习框架
实验环境
- Python 3.6.2
- PyTorch 1.7.1
- Flask 1.1.1
- Numpy 1.18.5
- Opencv 3.4.2
- PIL
pip3 install pillow
项目结构
相关说明：
- static：用于存储静态文件，比如css、js和图片等
- templates：存放模板文件
- upload：用于保存上传文件
- flask_app.py： 应用程序主文件
- predict.py：预测文件
主要代码
完整代码，暂时没空整理，如整理完，后续会发布，敬请期待！
#!/usr/bin/python# -*- coding: UTF-8 -*-import impfrom flask import request, jsonify, send_from_directory, abortfrom werkzeug.utils import secure_filenamefrom flask import Flask, render_template, jsonify, requestfrom predict import preimport timeimport osimport base64app = Flask(__name__)UPLOAD_FOLDER = 'upload'app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDERbasedir = os.path.abspath(os.path.dirname(__file__))ALLOWED_EXTENSIONS = set(['txt', 'png', 'jpg', 'xls', 'JPG', 'PNG', 'xlsx', 'gif', 'GIF'])# 用于判断文件后缀def allowed_file(filename): return '.' in filename and filename.rsplit('.', 1)[1] in ALLOWED_EXTENSIONS# 上传@app.route('/upload')def upload_test(): return render_template('upload.html')@app.route("/api/download/", methods=['GET'])def download(filename): if request.method == "GET": if os.path.isfile(os.path.join('upload', filename)): return send_from_directory('upload', filename, as_attachment=True) abort(404)# 上传文件@app.route('/api/upload', methods=['POST'], strict_slashes=False)def api_upload(): file_dir = os.path.join(basedir, app.config['UPLOAD_FOLDER']) if not os.path.exists(file_dir): os.makedirs(file_dir) f = request.files['myfile'] # 从表单的file字段获取文件，myfile为该表单的name值 if f and allowed_file(f.filename): # 判断是否是允许上传的文件类型 fname = secure_filename(f.filename) print(fname) ext = fname.rsplit('.', 1)[1] # 获取文件后缀 unix_time = int(time.time()) new_filename = str(unix_time) + '.' + ext # 修改了上传的文件名 f.save(os.path.join(file_dir, new_filename)) # 保存文件到upload目录 img_path = os.path.join("upload", new_filename) print(img_path) pre_result = pre(img_path) print(pre_result) token = base64.b64encode(new_filename.encode('utf-8')) print(token) return jsonify({"code": 0, "errmsg": "OK", "token": token, "fileName": "/api/download/" + new_filename,"detect_result:":pre_result}) else: return jsonify({"code": 1001, "errmsg": "ERROR"})if __name__ == '__main__':app.run(host="0.0.0.0",port="5000",threaded=True,debug=False)<!DOCTYPE html><html><head><meta charset="UTF-8"><link href="{{url_for('static', filename='obj_classification.css')}}" rel="stylesheet" type="text/css" /><title>图片识别--Person</title></head><body><h1>图片识别--Person</h1><div class="container"><div class="choose"><form action="http://IP地址:5000/api/upload" enctype='multipart/form-data' method='POST'><input type="file" name="myfile" class="input-new" style="margin-top:20px;" /><input type="submit" value="识别图片" class="button-new" style="margin-top:15px;" /></form></div><div class="display"><img src="{{ url_for('static', filename='images/test.jpg',_t=val1) }}" width="400" height="500" alt="图片" /></div></div></body></html>运行结果
{ "code": 0, "detect_result:": [ { "bbox": [ 51.0, 265.0, 543.0, 437.0 ], "class": "b'person 0.78'" }, { "bbox": [ 43.0, 433.0, 543.0, 609.0 ], "class": "b'person 0.77'" }, { "bbox": [ 44.0, 133.0, 543.0, 309.0 ], "class": "b'person 0.76'" }, { "bbox": [ 46.0, 526.0, 543.0, 665.0 ], "class": "b'person 0.74'" }, { "bbox": [ 107.0, 51.0, 525.0, 181.0 ], "class": "b'person 0.62'" } ], "errmsg": "OK", "fileName": "/api/download/1645974252.jpg", "token": "MTY0NTk3NDI1Mi5qcGc="}
参考链接
https://flask.palletsprojects.com/en/2.0.x/
