how to do image recognition in iphone apps

Question

Possible Duplicate:
An iPhone library for shape recognition via the camera

I am iPhone mobile apps developer and am looking for image recognition libraries to integrate in one of my App which should work like google's goggle functionality .

I saw https://www.iqengines.com/ and downlaoded its ios sdk and it worked well . But its not free version . I am searching for open source sdk .

also how opencv useful in accomplishing this ? beacuse some of the sources redirecting to openCV .

Please let me know if any one come across this . Thanks

Solution

Yes, OpenCV works on iOS and will provide you with a good library of tools to use. You can either build your own .framework (a bit tiresome) or download one of the internet.

• To build one, refer to this OpenCV guide
• If you want to download a pre-built framework, head over there

After that, you should be able to build computer vision software on iOS, be careful though, image processing can take a lot of power and memory.

OpenCV has its own C++ classes for images, you will probably need to convert them back and forth to UIImage for input and displaying

I leave here this piece of code I use for NSImage, you should find enough for you to change a bit the code for UIImage

//
//  NSImage+OpenCV.h
//

#import <AppKit/AppKit.h>

@interface NSImage (NSImage_OpenCV) {

} 

+(NSImage*)imageWithCVMat:(const cv::Mat&)cvMat;
-(id)initWithCVMat:(const cv::Mat&)cvMat;

@property(nonatomic, readonly) cv::Mat CVMat;
@property(nonatomic, readonly) cv::Mat CVGrayscaleMat;

@end

And

//
//  NSImage+OpenCV.mm
//

#import "NSImage+OpenCV.h"

static void ProviderReleaseDataNOP(void *info, const void *data, size_t size)
{
    return;
}


@implementation NSImage (NSImage_OpenCV)

-(CGImageRef)CGImage
{
    CGContextRef bitmapCtx = CGBitmapContextCreate(NULL/*data - pass NULL to let CG allocate the memory*/, 
                                                   [self size].width,  
                                                   [self size].height, 
                                                   8 /*bitsPerComponent*/, 
                                                   0 /*bytesPerRow - CG will calculate it for you if it's allocating the data.  This might get padded out a bit for better alignment*/, 
                                                   [[NSColorSpace genericRGBColorSpace] CGColorSpace], 
                                                   kCGBitmapByteOrder32Host|kCGImageAlphaPremultipliedFirst);

    [NSGraphicsContext saveGraphicsState];
    [NSGraphicsContext setCurrentContext:[NSGraphicsContext graphicsContextWithGraphicsPort:bitmapCtx flipped:NO]];
    [self drawInRect:NSMakeRect(0,0, [self size].width, [self size].height) fromRect:NSZeroRect operation:NSCompositeCopy fraction:1.0];
    [NSGraphicsContext restoreGraphicsState];

    CGImageRef cgImage = CGBitmapContextCreateImage(bitmapCtx);
    CGContextRelease(bitmapCtx);

    return cgImage;
}


-(cv::Mat)CVMat
{
    CGImageRef imageRef = [self CGImage];
    CGColorSpaceRef colorSpace = CGImageGetColorSpace(imageRef);
    CGFloat cols = self.size.width;
    CGFloat rows = self.size.height;
    cv::Mat cvMat(rows, cols, CV_8UC4); // 8 bits per component, 4 channels

    CGContextRef contextRef = CGBitmapContextCreate(cvMat.data,                 // Pointer to backing data
                                                    cols,                      // Width of bitmap
                                                    rows,                     // Height of bitmap
                                                    8,                          // Bits per component
                                                    cvMat.step[0],              // Bytes per row
                                                    colorSpace,                 // Colorspace
                                                    kCGImageAlphaNoneSkipLast |
                                                    kCGBitmapByteOrderDefault); // Bitmap info flags

    CGContextDrawImage(contextRef, CGRectMake(0, 0, cols, rows), imageRef);
    CGContextRelease(contextRef);
    CGImageRelease(imageRef);
    return cvMat;
}

-(cv::Mat)CVGrayscaleMat
{
    CGImageRef imageRef = [self CGImage];
    CGColorSpaceRef colorSpace = CGColorSpaceCreateDeviceGray();
    CGFloat cols = self.size.width;
    CGFloat rows = self.size.height;
    cv::Mat cvMat = cv::Mat(rows, cols, CV_8UC1); // 8 bits per component, 1 channel
    CGContextRef contextRef = CGBitmapContextCreate(cvMat.data,                 // Pointer to backing data
                                                    cols,                      // Width of bitmap
                                                    rows,                     // Height of bitmap
                                                    8,                          // Bits per component
                                                    cvMat.step[0],              // Bytes per row
                                                    colorSpace,                 // Colorspace
                                                    kCGImageAlphaNone |
                                                    kCGBitmapByteOrderDefault); // Bitmap info flags

    CGContextDrawImage(contextRef, CGRectMake(0, 0, cols, rows), imageRef);
    CGContextRelease(contextRef);
    CGColorSpaceRelease(colorSpace);
    CGImageRelease(imageRef);
    return cvMat;
}

+ (NSImage *)imageWithCVMat:(const cv::Mat&)cvMat
{
    return [[[NSImage alloc] initWithCVMat:cvMat] autorelease];
}

- (id)initWithCVMat:(const cv::Mat&)cvMat
{
    NSData *data = [NSData dataWithBytes:cvMat.data length:cvMat.elemSize() * cvMat.total()];

    CGColorSpaceRef colorSpace;

    if (cvMat.elemSize() == 1)
    {
        colorSpace = CGColorSpaceCreateDeviceGray();
    }
    else
    {
        colorSpace = CGColorSpaceCreateDeviceRGB();
    }

    CGDataProviderRef provider = CGDataProviderCreateWithCFData((__bridge CFDataRef)data);

    CGImageRef imageRef = CGImageCreate(cvMat.cols,                                     // Width
                                        cvMat.rows,                                     // Height
                                        8,                                              // Bits per component
                                        8 * cvMat.elemSize(),                           // Bits per pixel
                                        cvMat.step[0],                                  // Bytes per row
                                        colorSpace,                                     // Colorspace
                                        kCGImageAlphaNone | kCGBitmapByteOrderDefault,  // Bitmap info flags
                                        provider,                                       // CGDataProviderRef
                                        NULL,                                           // Decode
                                        false,                                          // Should interpolate
                                        kCGRenderingIntentDefault);                     // Intent   


    NSBitmapImageRep *bitmapRep = [[NSBitmapImageRep alloc] initWithCGImage:imageRef];
    NSImage *image = [[NSImage alloc] init];
    [image addRepresentation:bitmapRep];

    CGImageRelease(imageRef);
    CGDataProviderRelease(provider);
    CGColorSpaceRelease(colorSpace);

    return image;
}

@end

(source)