C#Winfrom和PYTHON接入騰訊雲OCR
阿新 • • 發佈：2018-11-20
程式基本目的是實現提取圖片中的文字。做了一個winfrom介面，接了騰訊雲的OCR圖片識別。
這裡接入文件就不給出了，想看的直接去騰訊雲看接入文件即可。
由於想做跨平臺的應用。分析了接入文件，並且之前想做一個web介面介面的，但是無奈javascript技術不到家，表單傳遞時的header頭問題遲遲沒有解決。
所有就在我的網站里弄了一個介面來計算我的Authorization，計算方法是直接採用騰訊給出的標準文件中的方法。直接把計算結果echo到web上。
而我windows上採用的是C#程式設計。主要想法就是先使用HTTPREQUEST方法做一個簡單的爬蟲。具體方法在我之前的部落格《C#使用Winfrom程式設計時的注意事項》有講過到。具體連結: https://blog.csdn.net/qq_27180763/article/details/82949365
主要還是卡在了header傳值的問題。由於Authorization不是標準Header值，需要採用別的方式來進行傳遞。這是比較需要注意的。
這個地方我主要採用了Content-Type=application/x-www-form-urlencoded,用POST方法把上傳檔案到騰訊雲的API介面，然後接受返回值而已。
接受的時候也要注意，由於收到的資訊是JSON碼的形式，需要自己下載一個Newtonsoft.Json.dll的檔案，並加入解決方案的引用當中。然後名稱空間引用
using Newtonsoft.Json; using Newtonsoft.Json.Linq;
這樣就可以直接使用JObject物件來轉換string型的json物件。並採用json標準來進行物件的訪問。
C# Winfrom的介面就不具體放出，我先給出C#的程式碼。
Windows下的程式碼
using Newtonsoft.Json.Linq; using System; using System.Collections.Generic; using System.IO; using System.Linq; using System.Net; using System.Text; using System.Threading; using System.Windows.Forms; namespace PDF轉WORD { public partial class Form2 : Form { private string filename=""; public Form2() { InitializeComponent(); System.Windows.Forms.Control.CheckForIllegalCrossThreadCalls = false; this.Text = "圖片轉PDF"; } private void button1_Click(object sender, EventArgs e) { OpenFileDialog fileDialog = new OpenFileDialog(); fileDialog.Multiselect = true; fileDialog.Title = "請選擇檔案"; fileDialog.Filter = "(*jpg*)|*.jpg*||(*.jpg*)||*.*"; if (fileDialog.ShowDialog() == DialogResult.OK) { string file = fileDialog.FileName; filename = file; label2.Text = filename.ToString(); MessageBox.Show("選擇成功！"); } } private void POST_DATA() { HttpWebRequest req = (HttpWebRequest)WebRequest.Create(new Uri(@"我的web伺服器")); req.Method = "POST"; req.UserAgent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:62.0) Gecko/20100101 Firefox/62.0"; req.ContentType = "application/x-www-form-urlencoded"; req.Accept = "application/json"; string result1 = ""; StringBuilder builder = new StringBuilder(); byte[] data = Encoding.UTF8.GetBytes(builder.ToString()); req.ContentLength = data.Length; using (Stream reqStream = req.GetRequestStream()) { reqStream.Write(data, 0, data.Length); reqStream.Close(); } HttpWebResponse resp = (HttpWebResponse)req.GetResponse(); Stream stream = resp.GetResponseStream(); //獲取響應內容 using (StreamReader reader = new StreamReader(stream, Encoding.UTF8)) { result1 = reader.ReadToEnd(); } HttpWebRequest webReq = (HttpWebRequest)WebRequest.Create(new Uri("http://recognition.image.myqcloud.com/ocr/general")); Stream memStream = new MemoryStream(); webReq.Method = "POST"; string boundary = "--------------" + DateTime.Now.Ticks.ToString("x");// 邊界符 webReq.ContentType = "multipart/form-data; boundary=" + boundary; byte[] enter = Encoding.ASCII.GetBytes("\r\n"); //換行 memStream.Write(enter, 0, enter.Length); Dictionary<string, string> dic = new Dictionary<string, string>() { {"appid","你的APPID"} }; //寫入文字欄位 string inputPartHeaderFormat = "--" + boundary + "\r\n" + "Content-Disposition:form-data;name=\"{0}\";" + "\r\n\r\n{1}\r\n"; foreach (var kv in dic) { string inputPartHeader = string.Format(inputPartHeaderFormat, kv.Key, kv.Value); var inputPartHeaderBytes = Encoding.ASCII.GetBytes(inputPartHeader); memStream.Write(inputPartHeaderBytes, 0, inputPartHeaderBytes.Length); } var fileStream = new FileStream(filename, FileMode.Open, FileAccess.Read); // 寫入檔案 string imagePartHeader = "--" + boundary + "\r\n" + "Content-Disposition: form-data; name=\"{0}\"; filename=\"{1}\"\r\n" + "Content-Type: image/jpeg\r\n\r\n"; var header = string.Format(imagePartHeader, "image", "1.jpg"); var headerbytes = Encoding.UTF8.GetBytes(header); memStream.Write(headerbytes, 0, headerbytes.Length); var buffer = new byte[1024]; int bytesRead; while ((bytesRead = fileStream.Read(buffer, 0, buffer.Length)) != 0) { memStream.Write(buffer, 0, bytesRead); } byte[] endBoundary = Encoding.ASCII.GetBytes("xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" + "\r\n" + boundary + "--\r\n"); memStream.Write(endBoundary, 0, endBoundary.Length); webReq.ContentLength = memStream.Length; webReq.Headers.Add(HttpRequestHeader.Authorization, result1); webReq.Host = "recognition.image.myqcloud.com"; var requestStream = webReq.GetRequestStream(); memStream.Position = 0; memStream.CopyTo(requestStream); HttpWebResponse response = (HttpWebResponse)webReq.GetResponse(); StreamReader sr = new StreamReader(response.GetResponseStream(), Encoding.UTF8); var ret = sr.ReadToEnd(); sr.Close(); response.Close(); requestStream.Close(); memStream.Close(); JObject content_info = JObject.Parse(ret); //這裡就是主要的對返回的json的處理了 for (int i = 0; i < content_info["data"]["items"].Count(); i++) { try { richTextBox1.AppendText(content_info["data"]["items"][i]["itemstring"].ToString()); } catch (Exception e) { MessageBox.Show(e.ToString()); } } } private void button2_Click(object sender, EventArgs e) { if(filename!="") { Thread th1 = new Thread(POST_DATA); th1.IsBackground = true; th1.Start(); } else { MessageBox.Show("請先選擇圖片。", "提示！"); } } private void button3_Click(object sender, EventArgs e) { filename = ""; if (filename == "") { label2.Text = "暫未選擇圖片"; MessageBox.Show("清空成功！"); } } private void button4_Click(object sender, EventArgs e) { richTextBox1.Clear(); } } }
以上就是主要介面的程式碼。而主介面的程式碼就不再放出，我先給出windows下的實驗效果。（照片帶水印）
Windows下的效果
LINUX上的程式碼實現（ubuntu）
用python實現的，所以程式碼比較簡潔。主要用到的函式庫為requests
沒什麼需要注意的==畢竟python比較簡單。直接安裝官方給的文件設定image和header就可以了。話不多說直接上程式碼QWQ
#!/usr/bin/env python
#encoding=utf-8
import re
import sys,os
import requests
def GetAuthorization():
html = requests.post(url="我的WEB伺服器地址")
html = html.text
info = html.split("\n")
return info[1].replace("\n","")
Authorization = str(GetAuthorization().replace("\n","").replace("\r",""))
def createRequest(filepath):
global Authorization
headers = {
'host':'recognition.image.myqcloud.com',
'Authorization':Authorization
}
files = {
'appid':(None,'你的appid'),
'image':('1.jpg',open(filepath,'rb'),'image/jpeg')
}
r = requests.post("http://recognition.image.myqcloud.com/ocr/general", files=files,headers=headers)
responseinfo = r.content
data = responseinfo.decode('utf-8')
print data
r_index = r'itemstring":"(.*?)"'
result = re.findall(r_index, data)
for i in result:
print(i)
def main():
createRequest(sys.argv[1])
if __name__ == '__main__':
main()
程式的第一個引數就是你要上傳檔案的絕對路徑==就是一個普通指令碼，沒有什麼很大的技術難度。
我要處理的照片還是上面那張，現在給出掃描結果。
LINUX下的效果圖
