using Newtonsoft.Json;
using System;
using System.Collections.Generic;
using System.Drawing;
using System.Drawing.Imaging;
using System.IO;
using System.Linq;
using System.Net;
using System.Net.Security;
using System.Security.Cryptography.X509Certificates;
using System.Text;
namespace Common.Utils
{
public static class OCRUtil
{
private const String Host = "https://ocrapi-document.taobao.com";
private const String Path = "/ocrservice/document";
private const String Method = "POST";
private const String Appcode = "************";
private static readonly IList<string> ImageExList = new List<string>()
{
".jpg",
".png",
".bmp"
};
/// <summary>
/// 确保图片可用
/// </summary>
/// <param name="fileName"></param>
private static void EnsureImageCanUse(string fileName)
{
if (string.IsNullOrWhiteSpace(fileName))
{
throw new ArgumentNullException(nameof(fileName), "文件名为空");
}
if (!File.Exists(fileName))
{
throw new FileNotFoundException("文件不存在", fileName);
}
string ex = System.IO.Path.GetExtension(fileName);
if (ImageExList.IndexOf(ex.ToLower()) == -)
{
throw new FileFormatException("图片格式需为jpg，png，bmp");
}
FileInfo fileInfo = new FileInfo(fileName);
if (fileInfo.Length > * * )
{
throw new Exception("文件不能大于4M");
}
}
/// <summary>
/// 将图片转换为Base64字符串
/// </summary>
/// <param name="fileName"></param>
/// <returns></returns>
private static string ImageToBase64String(string fileName)
{
Bitmap bitmap = new Bitmap(fileName);
string ex = System.IO.Path.GetExtension(fileName).ToLower();
using (MemoryStream ms = new MemoryStream())
{
ImageFormat format;
switch (ex)
{
case ".png":
format = ImageFormat.Png;
break;
case ".jpg":
format = ImageFormat.Jpeg;
break;
default:
format = ImageFormat.Bmp;
break;
}
bitmap.Save(ms, format);
byte[] arr = new byte[ms.Length];
ms.Position = ;
ms.Read(arr, , (int)ms.Length);
ms.Close();
return Convert.ToBase64String(arr);
}
}
/// <summary>
/// 阿里云OCR图片转文字
/// </summary>
/// <param name="fileName"></param>
/// <returns></returns>
public static string ImageToText(string fileName)
{
EnsureImageCanUse(fileName);
String querys = "";
String bodys = "{\"img\":\"" + ImageToBase64String(fileName) + "\",\"url\":\"\",\"prob\":false}";
String url = Host + Path;
HttpWebRequest httpRequest = null;
HttpWebResponse httpResponse = null;
if ( < querys.Length)
{
url = url + "?" + querys;
}
if (Host.Contains("https://"))
{
ServicePointManager.ServerCertificateValidationCallback = new RemoteCertificateValidationCallback(CheckValidationResult);
httpRequest = (HttpWebRequest)WebRequest.CreateDefault(new Uri(url));
}
else
{
httpRequest = (HttpWebRequest)WebRequest.Create(url);
}
httpRequest.Method = Method;
httpRequest.Headers.Add("Authorization", "APPCODE " + Appcode);
//根据API的要求，定义相对应的Content-Type
httpRequest.ContentType = "application/json; charset=UTF-8";
if ( < bodys.Length)
{
byte[] data = Encoding.UTF8.GetBytes(bodys);
using (Stream stream = httpRequest.GetRequestStream())
{
stream.Write(data, , data.Length);
}
}
httpResponse = (HttpWebResponse)httpRequest.GetResponse();
if (httpResponse.StatusCode != HttpStatusCode.OK)
{
throw new WebException("阿里云OCR接口调用识别失败");
}
Stream st = httpResponse.GetResponseStream();
StreamReader reader = new StreamReader(st, Encoding.GetEncoding("utf-8"));
string responseStr = reader.ReadToEnd();
OcrResult ocrResult = JsonConvert.DeserializeObject<OcrResult>(responseStr);
return OcrResultToString(ocrResult);
}
private static string OcrResultToString(OcrResult ocrResult)
{
StringBuilder sb = new StringBuilder();
var wordList = ocrResult.prism_wordsInfo;
foreach (var item in wordList)
{
int leftX = item.pos[].x;
int blankSpaceCount = (int)Math.Floor((double)leftX / );
if (blankSpaceCount > )
{
sb.Append(' ', blankSpaceCount*);
}
sb.AppendLine(item.word);
}
return sb.ToString();
}
public static bool CheckValidationResult(object sender, X509Certificate certificate, X509Chain chain, SslPolicyErrors errors)
{
return true;
}
}
/// <summary>
/// OCR识别结果
/// </summary>
public class OcrResult
{
public string sid { get; set; }
public string prism_version { get; set; }
public int prism_wnum { get; set; }
public List<Prism_Wordsinfo> prism_wordsInfo { get; set; }
public int height { get; set; }
public int width { get; set; }
public int orgHeight { get; set; }
public int orgWidth { get; set; }
}
public class Prism_Wordsinfo
{
public string word { get; set; }
public IList<Pos> pos { get; set; }
}
public class Pos
{
public int x { get; set; }
public int y { get; set; }
}
}