之前写检查报告单ocr识别只是脚本运行的方式，如何提供别人方便的使用，那就需要提供一个接口服务了。利用FlasK实现简单图片上传，将识别后的结果返回，并打包成docker镜像运行在k8s集群中，就可以方便对外服务了，在代码有一点改动但核心不变。

Flask 实现代码

app = Flask(__name__)
# 设置图片保存文件夹
UPLOAD_FOLDER = '/app/photo'
# UPLOAD_FOLDER = 'resizephoto'
app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
app.config['RESIZE_UPLOAD_FOLDER'] = '/app/resizephoto'

# 设置允许上传的文件格式
ALLOW_EXTENSIONS = ['png', 'jpg', 'jpeg']

# 判断文件后缀是否在列表中
def allowed_file(filename):
    return '.' in filename and filename.rsplit('.', 1)[-1] in ALLOW_EXTENSIONS

@app.route("/photo/upload", methods=['POST', "GET"])
def upload():
    if request.method == 'POST':
        # 获取post过来的文件名称，从name=file参数中获取
        file = request.files['file']
        yyname = request.form.get('yyname')
        resp = {}
        if file and allowed_file(file.filename):
            print(file.filename)
            # secure_filename方法会去掉文件名中的中文
            file_name = secure_filename(file.filename)
            # 保存图片
            if not os.path.exists(app.config['UPLOAD_FOLDER']):
                print("建立目录:",app.config['UPLOAD_FOLDER'])
                os.makedirs(app.config['UPLOAD_FOLDER'])
            if not os.path.exists(app.config['RESIZE_UPLOAD_FOLDER']):
                print("建立目录:", app.config['RESIZE_UPLOAD_FOLDER'])
                os.makedirs(app.config['RESIZE_UPLOAD_FOLDER'])
            uploadImagePath = os.path.join(app.config['UPLOAD_FOLDER'], file_name)
            file.save(uploadImagePath)
            resizeuploadImagePath = os.path.join(app.config['RESIZE_UPLOAD_FOLDER'])
            resizeimage = resizeimgae(resizeuploadImagePath, uploadImagePath) # 修改尺寸代码

            if yyname == "zl":
                nametmp = nameslistzl
            if yyname == "zyy":
                nametmp = nameslistzyy
            resdict = ocrsdk(resizeimage, nametmp) # 识别代码
            resp["data"] = resdict
            return json.dumps(resp,ensure_ascii=False),200
        else:
            resp["msg"] = "格式错误，请上传jpg格式文件"
            return json.dumps(resp,ensure_ascii=False),400
    return render_template('index.html')

# 查看图片
@app.route("/photo/.jpg")
def get_frame(imageId):
    # 图片上传保存的路径
    with open(r'D:\pythonProject\autoTestYek\baiduAIOCR\photo\{}.jpg'.format(imageId), 'rb') as f:
        image = f.read()
        resp = Response(image, mimetype="image/jpg")
        return resp

if __name__ == "__main__":
    app.run(host='0.0.0.0', port=5000, debug=True)

依赖包文件

requirements.txt

pandas==0.24.2
Pillow==9.2.0
baidu-aip
flask==1.1.2
chardet
jinja2==2.11.2
markupsafe==1.1.1
itsdangerous==1.1.0
werkzeug==1.0.1

Dockerfile

FROM python:3.7
RUN mkdir -pv /app
WORKDIR /app
COPY ./src  /app/src
# 安装python并配置环境
RUN  pip3 config set global.index-url https://mirrors.aliyun.com/pypi/simple/ \
        && pip3 install --upgrade pip
RUN pip3 install -r /app/src/requirements.txt
EXPOSE 5000
CMD ["python", "/app/src/main.py"]

运行后的镜像服务，部署在rancher中

html模板文件

  
  

请上传图片文件,并识别


   
医院名称(默认长沙市中医医院)

文件结构

实现效果

可以用Postman测试验证，也可以页面请求测试

页面效果

测试请求

页面请求测试：

postman测试：

返回结果