Revision history [back]

Scan and find the length and breadth of object from iPhone camera

I want to find the length, breadth and outside diameter of the object

My analysis so far:

Applied the multiple algorithms like Canny and Sobel to detect the edges of the object.

I converted a image into Matrix Array:

//Function that convert image to matrix array. - (cv::Mat)cvMatFromUIImage:(UIImage *)image { CGColorSpaceRef colorSpace = CGImageGetColorSpace(image.CGImage); CGFloat cols = image.size.width; CGFloat rows = image.size.height;

cv::Mat cvMat(rows, cols, CV_8UC4); // 8 bits per component, 4 channels (color channels + alpha)

CGContextRef contextRef = CGBitmapContextCreate(cvMat.data,                 // Pointer to  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), image.CGImage);
CGContextRelease(contextRef);

return cvMat;

}

//Use function to create a matrix object.

cv::Mat cvImage = [self cvMatFromUIImage:image];

Then I apply the canny:

cv::Canny(grayImage, edges, m_cannyLoThreshold, m_cannyHiThreshold, m_cannyAperture * 2 + 1);

Same way I applied Sobel algorithm as well:

int scale = 1;
int delta = 0;
int ddepth = CV_16S;

/// Gradient X
cv::Sobel( grayImage, grad_x, ddepth, 1, 0, 3, scale, delta, cv::BORDER_DEFAULT );   
cv::convertScaleAbs( grad_x, abs_grad_x );

/// Gradient Y  
cv::Sobel( grayImage, grad_y, ddepth, 0, 1, 3, scale, delta, cv::BORDER_DEFAULT );   
cv::convertScaleAbs( grad_y, abs_grad_y );

/// Total Gradient (approximate)
cv::addWeighted( abs_grad_x, 0.5, abs_grad_y, 0.5, 0, edges );

Now, I am separating the black and white pixels on the image. But every pixels that is looking black on device return some other pixel color.

What I have to do, separating the pixels from the image so that I get the actual height and width of object in image. Any suggestion ?

I have to detect the object in below image. image description