I am currently having problems with CGImageRef.
Whenever I create a CGImageRef and look at it in debugger view, in Xcode, it is nil.
Here's the code:
-(void)mouseMoved:(NSEvent *)theEvent{
if (self.shoulddrag) {
NSPoint event_location = [theEvent locationInWindow];//direct from the docs
NSPoint local_point = [self convertPoint:event_location fromView:nil];//direct from the docs
CGImageRef theImage = (__bridge CGImageRef)(self.image);
CGImageRef theClippedImage = CGImageCreateWithImageInRect(theImage, CGRectMake(local_point.x,local_point.y,1,1));
NSImage * image = [[NSImage alloc] initWithCGImage:theClippedImage size:NSZeroSize];
self.pixleView.image = image;
CGImageRelease(theClippedImage);
}
}
Everything else seems to be working though. I can't understand. Any help would be appreciated.
Note: self.pixelView is an NSImageView instance that has not been overridden in any way.
Very likely local_point is not inside of the image. You've converted the point from the window to the view coordinates, but that may not be equivalent to the image coordinates. Test this to see if the lower-left corner of your image results in local_point being (0,0).
It's not clear how your view is laid out, but I suspect that what you want to do is subtract the origin of whatever region (possibly a subview) the user is interacting with relative to self.
Alright, I figured it out.
What I was using to create the CGImageRef was:
CGImageRef theImage = (__bridge CGImageRef)(self.image);
Apparently what I should have used is:
CGImageSourceRef theImage = CGImageSourceCreateWithData((CFDataRef)[self.image TIFFRepresentation], NULL);
I guess my problem was that for some reason I thought NSImage and CGImageRef had toll free bridging.
Apparently, I was wrong.
Related
I have a CGImageRef (lets call it original image) and a transparent png (watermark). I'm trying to write a method to place the watermark on top of the original, and return a CGImageRef.
In iOS I would have used UIKit to draw them both onto a context, but that doesn't seem possible with OSX (doesn't support UIKit).
Whats the simplest way to stack two images? Thanks
For a quick 'n dirty solution you can use the NSImage drawing APIs:
NSImage *background = [NSImage imageNamed:#"background"];
NSImage *overlay = [NSImage imageNamed:#"overlay"];
NSImage *newImage = [[NSImage alloc] initWithSize:[background size]];
[newImage lockFocus];
CGRect newImageRect = CGRectZero;
newImageRect.size = [newImage size];
[background drawInRect:newImageRect];
[overlay drawInRect:newImageRect];
[newImage unlockFocus];
CGImageRef newImageRef = [newImage CGImageForProposedRect:NULL context:nil hints:nil];
If you don't like that, most of the CGContext APIs you'd expect are available cross platform—for drawing with a little more control. Similarly, you could look into NSGraphicsContext.
This is pretty easy when you render to a CGContext.
If you want an image as a result, you can create and render to a CGBitmapContext, then request the image after render.
General flow, with common details and contextual info omitted:
CGImageRef CreateCompositeOfImages(CGImageRef pBackground,
const CGRect pBackgroundRect,
CGImageRef pForeground,
const CGRect pForegroundRect)
{
// configure context parameters
CGContextRef gtx = CGBitmapContextCreate( %%% );
// configure context
// configure context to render background image
// draw background image
CGContextDrawImage(gtx, pBackgroundRect, pBackground);
// configure context to render foreground image
// draw foreground image
CGContextDrawImage(gtx, pForegroundRect, pForeground);
// create result
CGImageRef result = CGBitmapContextCreateImage(gtx);
// cleanup
return result;
}
You would need to create a CGImage from your PNG.
Additional APIs you may be interested in using:
CGContextSetBlendMode
CGContextSetAllowsAntialiasing
CGContextSetInterpolationQuality.
I know a lot of people will generally advise you to use higher level abstractions (i.e. AppKit and UIKit), but CoreGraphics is a great library for rendering in both of those contexts. If you are interested in graphics implementations which are easy to use in both OS X and iOS, CoreGraphics is a good choice to base your work upon if you are comfortable working with those abstractions.
If anyone, like me, needs a Swift version.
This is a functional Swift 5 version:
let background = NSImage(named: "background")
let overlay = NSImage(named: "overlay")
let newImage = NSImage(size: background.size)
newImage.lockFocus()
var newImageRect: CGRect = .zero
newImageRect.size = newImage.size
background.draw(in: newImageRect)
overlay.draw(in: newImageRect)
newImage.unlockFocus()
I wish I had the time to do the same with the CGContext example.
Is there any way in Objective-c/cocoa (OSX) to crop an image without changing the quality of the image?
I am very near to a solution, but there are still some differences that I can detect in the color. I can notice it when zooming into the text. Here is the code I am currently using:
NSImage *target = [[[NSImage alloc]initWithSize:panelRect.size] autorelease];
target.backgroundColor = [NSColor greenColor];
//start drawing on target
[target lockFocus];
[NSGraphicsContext saveGraphicsState];
[[NSGraphicsContext currentContext] setImageInterpolation:NSImageInterpolationNone];
[[NSGraphicsContext currentContext] setShouldAntialias:NO];
//draw the portion of the source image on target image
[source drawInRect:NSMakeRect(0,0,panelRect.size.width,panelRect.size.height)
fromRect:NSMakeRect(panelRect.origin.x , source.size.height - panelRect.origin.y - panelRect.size.height, panelRect.size.width, panelRect.size.height)
operation:NSCompositeCopy
fraction:1.0];
[NSGraphicsContext restoreGraphicsState];
//end drawing
[target unlockFocus];
//create a NSBitmapImageRep
NSBitmapImageRep *bmpImageRep = [[[NSBitmapImageRep alloc]initWithData:[target TIFFRepresentation]] autorelease];
//add the NSBitmapImage to the representation list of the target
[target addRepresentation:bmpImageRep];
//get the data from the representation
NSData *data = [bmpImageRep representationUsingType: NSJPEGFileType
properties: imageProps];
NSString *filename = [NSString stringWithFormat:#"%#%#.jpg", panelImagePrefix, panelNumber];
NSLog(#"This is the filename: %#", filename);
//write the data to a file
[data writeToFile:filename atomically:NO];
Here is a zoomed-in comparison of the original and the cropped image:
(Original image - above)
(Cropped image - above)
The difference is hard to see, but if you flick between them, you can notice it. You can use a colour picker to notice the difference as well. For example, the darkest pixel on the bottom row of the image is a different shade.
I also have a solution that works exactly the way I want it in iOS. Here is the code:
-(void)testMethod:(int)page forRect:(CGRect)rect{
NSString *filePath = #"imageName";
NSData *data = [HeavyResourceManager dataForPath:filePath];//this just gets the image as NSData
UIImage *image = [UIImage imageWithData:data];
CGImageRef imageRef = CGImageCreateWithImageInRect([image CGImage], rect);//crop in the rect
UIImage *result = [UIImage imageWithCGImage:imageRef scale:0 orientation:image.imageOrientation];
CGImageRelease(imageRef);
NSArray *paths = NSSearchPathForDirectoriesInDomains(NSDocumentDirectory, NSUserDomainMask, YES);
NSString *documentsDirectoryPath = [paths objectAtIndex:0];
[UIImageJPEGRepresentation(result, 1.0) writeToFile:[documentsDirectoryPath stringByAppendingPathComponent::#"output.jpg"] atomically:YES];
}
So, is there a way to crop an image in OSX so that the cropped image does not change at all? Perhaps I have to look into a different library, but I would be surprised if I could not do this with Objective-C...
Note, This is a follow up question to my previous question here.
Update I have tried (as per the suggestion) to round the CGRect values to whole numbers, but did not notice a difference. Here is the code in case I used:
[source drawInRect:NSMakeRect(0,0,(int)panelRect.size.width,(int)panelRect.size.height)
fromRect:NSMakeRect((int)panelRect.origin.x , (int)(source.size.height - panelRect.origin.y - panelRect.size.height), (int)panelRect.size.width, (int)panelRect.size.height)
operation:NSCompositeCopy
fraction:1.0];
Update I have tried mazzaroth code and it works if I save it as a png, but if I try and save it as a jpeg, the image loses quality. So close, but not close enough. Still hoping for a complete answer...
use CGImageCreateWithImageInRect.
// this chunk of code loads a jpeg image into a cgimage
// creates a second crop of the original image with CGImageCreateWithImageInRect
// writes the new cropped image to the desktop
// ensure that the xy origin of the CGRectMake call is smaller than the width or height of the original image
NSURL *originalImage = [NSURL fileURLWithPath:[[NSBundle mainBundle] pathForResource:#"lockwood" ofType:#"jpg"]];
CGImageRef imageRef = NULL;
CGImageSourceRef loadRef = CGImageSourceCreateWithURL((CFURLRef)originalImage, NULL);
if (loadRef != NULL)
{
imageRef = CGImageSourceCreateImageAtIndex(loadRef, 0, NULL);
CFRelease(loadRef); // Release CGImageSource reference
}
CGImageRef croppedImage = CGImageCreateWithImageInRect(imageRef, CGRectMake(200., 200., 100., 100.));
CFURLRef saveUrl = (CFURLRef)[NSURL fileURLWithPath:[#"~/Desktop/lockwood-crop.jpg" stringByExpandingTildeInPath]];
CGImageDestinationRef destination = CGImageDestinationCreateWithURL(saveUrl, kUTTypeJPEG, 1, NULL);
CGImageDestinationAddImage(destination, croppedImage, nil);
if (!CGImageDestinationFinalize(destination)) {
NSLog(#"Failed to write image to %#", saveUrl);
}
CFRelease(destination);
CFRelease(imageRef);
CFRelease(croppedImage);
I also made a gist:
https://gist.github.com/4259594
Try to change the drawInRect orign to 0.5,0.5. Otherwise Quartz will distribute each pixel color to the adjacent 4 fixels.
Set the color space of the target image. You might be having a different colorspace causing to to look slightly different.
Try the various rendering intents and see which gets the best result, perceptual versus relative colorimetric etc. There are 4 options I think.
You mention that the colors get modified by the saving of JPEG versus PNG.
You can specify the compression level when saving to JPEG. Try with something like 0.8 or 0.9. you can also save JPEG without compression with 1.0, but ther PNG has a distinct advantage. You specify the compression level in the options dictionary for CGImageDesinationAddImage.
Finally - if nothing her helps - you should open a TSI with DTS, they can certainly provide you with the guidance you seek.
The usual problem is that cropping sizes are float, but image pixels are integer.
cocoa interpolates it automatically.
You need to floor, round or ceil the size and coordinates to be sure that they are integer.
This may help.
I am doing EXIF deleting of JPG files and I think I caught the reason:
All losses and changes come from the re-compress of your image during saving to file.
You may notice the change too if you just to save the whole image again.
What I am to do is to read the original JPG and re-compress it to a quality that take equivalent file size.
I am trying to get the pixel colour from an image displayed by the webcam. I want to see how the pixel colour is changing with time.
My current solution sucks a LOT of CPU, it works and gives me the correct answer, but I am not 100% sure if I am doing this correctly or I could cut some steps out.
- (IBAction)addFrame:(id)sender
{
// Get the most recent frame
// This must be done in a #synchronized block because the delegate method that sets the most recent frame is not called on the main thread
CVImageBufferRef imageBuffer;
#synchronized (self) {
imageBuffer = CVBufferRetain(mCurrentImageBuffer);
}
if (imageBuffer) {
// Create an NSImage and add it to the movie
// I think I can remove some steps here, but not sure where.
NSCIImageRep *imageRep = [NSCIImageRep imageRepWithCIImage:[CIImage imageWithCVImageBuffer:imageBuffer]];
NSSize n = {320,160 };
//NSImage *image = [[[NSImage alloc] initWithSize:[imageRep size]] autorelease];
NSImage *image = [[[NSImage alloc] initWithSize:n] autorelease];
[image addRepresentation:imageRep];
CVBufferRelease(imageBuffer);
NSBitmapImageRep* raw_img = [NSBitmapImageRep imageRepWithData:[image TIFFRepresentation]];
NSLog(#"image width is %f", [image size].width);
NSColor* color = [raw_img colorAtX:1279 y:120];
float colourValue = [color greenComponent]+ [color redComponent]+ [color blueComponent];
[graphView setXY:10 andY:200*colourValue/3];
NSLog(#"%0.3f", colourValue);
Any help is appreciated and I am happy to try other ideas.
Thanks guys.
There are a couple of ways that this could be made more efficient. Take a look at the imageFromSampleBuffer: method in this Tech Q&A, which presents a cleaner way of getting from a CVImageBufferRef to an image (the sample uses a UIImage, but it's practically identical for an NSImage).
You can also pull the pixel values straight out of the CVImageBufferRef without any conversion. Once you have the base address of the buffer, you an calculate the offset of any pixel and just read the values from there.
I am trying to capture video from a camera. i have gotten the captureOutput:didOutputSampleBuffer: callback to trigger and it gives me a sample buffer that i then convert to a CVImageBufferRef. i then attempt to convert that image to a UIImage that i can then view in my app.
- (void)captureOutput:(AVCaptureOutput *)captureOutput didOutputSampleBuffer:(CMSampleBufferRef)sampleBuffer fromConnection:(AVCaptureConnection *)connection
{
CVImageBufferRef imageBuffer = CMSampleBufferGetImageBuffer(sampleBuffer);
/*Lock the image buffer*/
CVPixelBufferLockBaseAddress(imageBuffer,0);
/*Get information about the image*/
uint8_t *baseAddress = (uint8_t *)CVPixelBufferGetBaseAddress(imageBuffer);
size_t bytesPerRow = CVPixelBufferGetBytesPerRow(imageBuffer);
size_t width = CVPixelBufferGetWidth(imageBuffer);
size_t height = CVPixelBufferGetHeight(imageBuffer);
/*We unlock the image buffer*/
CVPixelBufferUnlockBaseAddress(imageBuffer,0);
/*Create a CGImageRef from the CVImageBufferRef*/
CGColorSpaceRef colorSpace = CGColorSpaceCreateDeviceRGB();
CGContextRef newContext = CGBitmapContextCreate(baseAddress, width, height, 8, bytesPerRow, colorSpace, kCGBitmapByteOrder32Little | kCGImageAlphaPremultipliedFirst);
CGImageRef newImage = CGBitmapContextCreateImage(newContext);
/*We release some components*/
CGContextRelease(newContext);
CGColorSpaceRelease(colorSpace);
/*We display the result on the custom layer*/
/*self.customLayer.contents = (id) newImage;*/
/*We display the result on the image view (We need to change the orientation of the image so that the video is displayed correctly)*/
UIImage *image= [UIImage imageWithCGImage:newImage scale:1.0 orientation:UIImageOrientationRight];
self.capturedView.image = image;
/*We relase the CGImageRef*/
CGImageRelease(newImage);
}
the code seems to work fine up until the call to CGBitmapContextCreate. it always returns a NULL pointer. so consequently none of the rest of the function works. no matter what i seem to pass it the function returns null. i have no idea why.
The way that you are passing on the baseAddress presumes that the image data is in the form
ACCC
( where C is some color component, R || G || B ).
If you've set up your AVCaptureSession to capture the video frames in native format, more than likely you're getting the video data back in planar YUV420 format. (see: link text ) In order to do what you're attempting to do here, probably the easiest thing to do would be specify that you want the video frames captured in kCVPixelFormatType_32RGBA . Apple recommends that you capture the video frames in kCVPixelFormatType_32BGRA if you capture it in non-planar format at all, the reasoning for which is not stated, but I can reasonably assume is due to performance considerations.
Caveat: I've not done this, and am assuming that accessing the CVPixelBufferRef contents like this is a reasonable way to build the image. I can't vouch for this actually working, but I /can/ tell you that the way you are doing things right now reliably will not work due to the pixel format that you are (probably) capturing the video frames as.
If you need to convert a CVImageBufferRef to UIImage, it seems to be much more difficult than it should be unfortunately.
Essentially you need to first convert it to CIImage, then CGImage, and then finally UIImage. I wish I could tell you why. :)
-(void) screenshotOfVideoStream:(CVImageBufferRef)imageBuffer
{
CIImage *ciImage = [CIImage imageWithCVPixelBuffer:imageBuffer];
CIContext *temporaryContext = [CIContext contextWithOptions:nil];
CGImageRef videoImage = [temporaryContext
createCGImage:ciImage
fromRect:CGRectMake(0, 0,
CVPixelBufferGetWidth(imageBuffer),
CVPixelBufferGetHeight(imageBuffer))];
UIImage *image = [[UIImage alloc] initWithCGImage:videoImage];
[self doSomethingWithOurUIImage:image];
CGImageRelease(videoImage);
}
This particular method worked for me when I was converting H.264 video using the VTDecompressionSession callback to get the CVImageBufferRef (but it should work for any CVImageBufferRef). I was using iOS 8.1, XCode 6.2.
You can directly call:
self.yourImageView.image=[[UIImage alloc] initWithCIImage:[CIImage imageWithCVPixelBuffer:imageBuffer]];
Benjamin Loulier wrote a really good post on outputting a CVImageBufferRef under the consideration of speed with multiple approaches.
You can also find a working example on github ;)
How about back in time? ;)
Here you go: http://web.archive.org/web/20140426162537/http://www.benjaminloulier.com/posts/ios4-and-direct-access-to-the-camera
Is there an easy way to do this that works in 10.5?
In 10.6 I can use nsImage CGImageForProposedRect: NULL context: NULL hints: NULL
If I'm not using 1b black and white images (Like Group 4 TIFF), I can use bitmaps, but cgbitmaps seem to not like that setup... Is there a general way of doing this?
I need to do this because I have an IKImageView that seems to only want to add CGImages, but all I've got are NSImages. Currently, I'm using a private setImage:(NSImage*) method that I'd REALLY REALLY rather not be using...
Found the following solution on this page:
NSImage* someImage;
// create the image somehow, load from file, draw into it...
CGImageSourceRef source;
source = CGImageSourceCreateWithData((CFDataRef)[someImage TIFFRepresentation], NULL);
CGImageRef maskRef = CGImageSourceCreateImageAtIndex(source, 0, NULL);
All the methods seem to be 10.4+ so you should be fine.
[This is the long way around. You should only do this if you're still supporting an old version of OS X. If you can require 10.6 or later, just use the CGImage method instead.]
Create a CGBitmapContext.
Create an NSGraphicsContext for the bitmap context.
Set the graphics context as the current context.
Draw the image.
Create the CGImage from the contents of the bitmap context.
Release the bitmap context.
As of Snow Leopard, you can just ask the image to create a CGImage for you, by sending the NSImage a CGImageForProposedRect:context:hints: message.
Here's a more detailed answer for using - CGImageForProposedRect:context:hints::
NSImage *image = [NSImage imageNamed:#"image.jpg"];
NSRect imageRect = NSMakeRect(0, 0, image.size.width, image.size.height);
CGImageRef cgImage = [image CGImageForProposedRect:&imageRect context:NULL hints:nil];
Just did this using CGImageForProposedRect:context:hints:...
NSImage *image; // an image
NSGraphicsContext *context = [NSGraphicsContext currentContext];
CGRect imageCGRect = CGRectMake(0, 0, image.size.width, image.size.height);
NSRect imageRect = NSRectFromCGRect(imageCGRect);
CGImageRef imageRef = [image CGImageForProposedRect:&imageRect context:context hints:nil];
As of 2021, CALayers contents can directly be feed with NSImage but you still may get an [Utility] unsupported surface format: LA08 warning.
After a couple of days research and testing around i found out that this Warning is triggered if you created an NSView with backingLayer, aka CALayer and just used an NSImage to feed its contents. This alone is not much of a problem. But if you try to render it via [self.layer renderInContext:ctx], each rendering will trigger the warning again.
To make use simple, i created an NSImage extension to fix this..
#interface NSImage (LA08FIXExtension)
-(id)CGImageRefID;
#end
#implementation NSImage (LA08FIXExtension)
-(id)CGImageRefID {
NSSize size = [self size];
NSRect rect = NSMakeRect(0, 0, size.width, size.height);
CGImageRef ref = [self CGImageForProposedRect:&rect context:[NSGraphicsContext currentContext] hints:NULL];
return (__bridge id)ref;
}
#end
see how it does not return an CGImageRef directly but a bridged cast to id..! This does the trick.
Now you can use it like..
CALayer *somelayer = [CALayer layer];
somelayer.frame = ...
somelayer.contents = [NSImage imageWithName:#"SomeImage"].CGImageRefID;
PS: posted her because if you got that problem also, google will lead you to this Thread of answers anyway, and all above answers inspired me for this fix.