分享一下我老師大神的人工智慧教程！零基礎，通俗易懂！http://blog.csdn.net/jiangjunshow

也歡迎大家轉載本篇文章。分享知識，造福人民，實現我們中華民族偉大復興！

今天寫demo的時候發現, 如果把通過相機獲取到的圖片，直接進行操作, 比如裁剪, 縮放, 則會把原圖片向又旋轉90度。

剛開始覺得莫名其妙, 不知所措。 後來百度了一下，找到了解決辦法。

ps: 查詢過程中, 碰到了一種說法:

//get original photo from iOS photos //如果該圖片大於2M，會自動旋轉90度；否則不旋轉UIImage* originalImg=[dict objectForKey:UIImagePickerControllerOriginalImage];

下面的解決辦法親測可行。 原文:http://www.cnblogs.com/jiangyazhou/archive/2012/03/22/2412343.html

用相機拍攝出來的照片含有EXIF資訊，UIImage的imageOrientation屬性指的就是EXIF中的orientation資訊。
如果我們忽略orientation資訊，而直接對照片進行畫素處理或者drawInRect等操作，得到的結果是翻轉或者旋轉90之後的樣子。這是因為我們執行畫素處理或者drawInRect等操作之後，imageOrientaion資訊被刪除了，imageOrientaion被重設為0，造成照片內容和imageOrientaion不匹配。
所以，在對照片進行處理之前，先將照片旋轉到正確的方向，並且返回的imageOrientaion為0。
下面這個方法就是一個UIImage category中的方法，用它可以達到以上目的。

- (UIImage *)fixOrientation:(UIImage *)aImage {        // No-op if the orientation is already correct    if (aImage.imageOrientation == UIImageOrientationUp)         return aImage;        // We need to calculate the proper transformation to make the image upright.
 
    // We do it in 2 steps: Rotate if Left/Right/Down, and then flip if Mirrored.    CGAffineTransform transform = CGAffineTransformIdentity;        switch (aImage.imageOrientation) {        case UIImageOrientationDown:        case UIImageOrientationDownMirrored:            transform = CGAffineTransformTranslate(transform, aImage.size.width, aImage.size.height);            transform = CGAffineTransformRotate(transform, M_PI);            break;                    case UIImageOrientationLeft:        case UIImageOrientationLeftMirrored:            transform = CGAffineTransformTranslate(transform, aImage.size.width, 0);            transform = CGAffineTransformRotate(transform, M_PI_2);            break;                    case UIImageOrientationRight:        case UIImageOrientationRightMirrored:            transform = CGAffineTransformTranslate(transform, 0, aImage.size.height);            transform = CGAffineTransformRotate(transform, -M_PI_2);            break;        default:            break;    }        switch (aImage.imageOrientation) {        case UIImageOrientationUpMirrored:        case UIImageOrientationDownMirrored:            transform = CGAffineTransformTranslate(transform, aImage.size.width, 0);            transform = CGAffineTransformScale(transform, -1, 1);            break;                    case UIImageOrientationLeftMirrored:        case UIImageOrientationRightMirrored:            transform = CGAffineTransformTranslate(transform, aImage.size.height, 0);            transform = CGAffineTransformScale(transform, -1, 1);            break;        default:            break;    }        // Now we draw the underlying CGImage into a new context, applying the transform    // calculated above.    CGContextRef ctx = CGBitmapContextCreate(NULL, aImage.size.width, aImage.size.height,                                             CGImageGetBitsPerComponent(aImage.CGImage), 0,                                             CGImageGetColorSpace(aImage.CGImage),                                             CGImageGetBitmapInfo(aImage.CGImage));    CGContextConcatCTM(ctx, transform);    switch (aImage.imageOrientation) {        case UIImageOrientationLeft:        case UIImageOrientationLeftMirrored:        case UIImageOrientationRight:        case UIImageOrientationRightMirrored:            // Grr...            CGContextDrawImage(ctx, CGRectMake(0,0,aImage.size.height,aImage.size.width), aImage.CGImage);            break;                    default:            CGContextDrawImage(ctx, CGRectMake(0,0,aImage.size.width,aImage.size.height), aImage.CGImage);            break;    }        // And now we just create a new UIImage from the drawing context    CGImageRef cgimg = CGBitmapContextCreateImage(ctx);    UIImage *img = [UIImage imageWithCGImage:cgimg];    CGContextRelease(ctx);    CGImageRelease(cgimg);    return img;}

給我老師的人工智慧教程打call！http://blog.csdn.net/jiangjunshow