I'm currently building a site in app engine that uploads images to google cloud storage and to complete basic manipulations I'm using python's PIL
I've been having problems with the following image which another stackoverflow member has mentioned is a palette-based PNG with transparency, which I've been reading may be a bit buggy in PIL
My question is really a back to basics one: What is the best way to convert this to an RGB format with transparent pixels set to #FFF? I've been able to get it to work through a combined RGBA then RGB paste but that seems redundant
However, for a direct conversion I'm getting a bad transparency mask i.e. using the solution from PIL Convert PNG or GIF with Transparency to JPG without
Also if anybody has ideas why the image degrades to terrible quality after conversion, that's entirely a bonus for me!
A way to do this is to first convert the file to jpg -- seems like a problem with the png encoding (or something related to that)
Check out this link that I used and got smooth conversion from transparent PNG to GIF:
Convert RGBA PNG to RGB with PIL
The function you are looking for is pure_pil_alpha_to_color_v2.
I also used for my image conversion tool PySmile:
https://github.com/vietlq/PySmile/blob/master/pysmile.py
Related
I dont understand the explanation about why GIF files always has image resolution 72ppi.
I have tried using ImageMagick converting PNG to GIF specifying -density to be 300ppi, but the resulting GIF is still 72ppi.
The GIF format does not store any value for "ppi" or "dpi" or pixel density/resolution - see Wikipedia.
I think software assumes 72 dpi, and then loads your image, updating that assumption if it finds anything better... which it never will in the case of GIF.
In any case, the dpi/ppi is irrelevant until such time as you want to print your image, and I don't know of anyone ever using GIF as a method of submission of images to any reputable printers or publications - TIFF is probably the preferred or most common method.
We are currently converting TIFF files to PDF using GraphicsMagick. The TIFF is coming from an eFAX and has a (pixel) resolution of 1728x2200.
If you do the conversion with tiff2pdf or just open it on Preview and convert export it to PDF, it is generated with a MediaBox value of 612x792 point, which is what is expected.
However graphics magick generates a MediaBox of 1728x4400 and a CropBox of 610x792. It all looks good if you open it on a PDF viewer because it's using the CropBox but if you're feeding it to GhostScript after, you don't get the Image on the full page but as a small square inside the document.
The lazy solutions would be to change for Tiff2PDF or add -dUseCropBox to our GhostScript command but I'd like to know what GraphicsMagick option should be used to have the PDF with the good MediaBox. It's like it doesn't understand that the resolution is in Pixels and not in Point. Hope somebody has insights
I am using OpenCV 3.0 and whenever I read an image and write it back the result is a washed-out image.
code:
cv::Mat img = cv::imread("dir/frogImage.jpg",-1);
cv::imwrite("dir/result.jpg",img);
Does anyone know whats causing this?
Original:
Result:
You can try to increase the compression quality parameter as shown in OpenCV Documentation of cv::imwrite :
cv::Mat img = cv::imread("dir/frogImage.jpg",-1);
std::vector<int> compression_params;
compression_params.push_back(CV_IMWRITE_JPEG_QUALITY);
compression_params.push_back(100);
cv::imwrite("dir/result.jpg",img, compression_params);
Without specifying the compression quality manually, quality of 95% will be applied.
but 1. you don't know what jpeg compression quality your original image had (so maybe you might increase the image size) and 2. it will (afaik) still introduce additional minor artifacts, because after all it is a lossy compression method.
UPDATE your problem seems to be not because of compression artifacts but because of an image with Adobe RGB 1998 color format. OpenCV interprets the color values as they are, but instead it should scale the color values to fit the "real" RGB color space. Browser and some image viewers do apply the color format correctly, while others don't (e.g. irfanView). I used GIMP to verify. Using GIMP you can decide on startup how to interpret the color values by format, either getting your desired or your "washed out" image.
OpenCV definitely doesn't care about such things, since it's not a photo editing library, so neither on reading nor on writing, color format will be handled.
This is because you are saving the image as JPG. When doing this the OpenCV will compress the image.
try to save it as PNG or BMP and no difference will be exist.
However, the IMPORTANT QUESTION : I am loading the image as jpg and saving it as JPG. So, how there is a difference?!
Yes, this is because there is many not identical compression/decompression algorithms for JPG.
if you want to get into some details see this question:
Reading jpg file in OpenCV vs C# Bitmap
EDIT:
You can see what I mean exactly here:
auto bmp(cv::imread("c:/Testing/stack.bmp"));
cv::imwrite("c:/Testing/stack_OpenCV.jpg", bmp);
auto jpg_opencv(cv::imread("c:/Testing/stack_OpenCV.jpg"));
auto jpg_mspaint(cv::imread("c:/Testing/stack_mspaint.jpg"));
cv::imwrite("c:/Testing/stack_mspaint_opencv.jpg", jpg_mspaint);
jpg_mspaint=(cv::imread("c:/Testing/stack_mspaint_opencv.jpg"));
cv::Mat jpg_diff;
cv::absdiff(jpg_mspaint, jpg_opencv, jpg_diff);
std::cout << cv::mean(jpg_diff);
The Result:
[0.576938, 0.466718, 0.495106, 0]
As #Micha commented:
cv::Mat img = cv::imread("dir/frogImage.jpg",-1);
cv::imwrite("dir/result.bmp",img);
I was always annoyed when mspaint.exe did the same to jpeg images. Especially for the screenshots...it ruined them everytime.
IPA image uses pngcrush to compress PNG image, but I want to uncrush a PNG image on Ubuntu.
Can anyone give me any idea?
The standard PNG utility pngcrush has been modified by Apple, which makes it produce technically invalid PNGs: a new chunk is inserted before the mandatory first chunk IHDR, RGB(A) order of pixel data is inverted, and RGB pixels get premultiplied with their alpha.
Hence, I'd rather call these PNGs "fried", rather than just "crushed".
Try my own pngdefry. The source code is written on a Mac OSX machine but it should be compilable for other OSes as well; it's pretty straightforward C code.
Here is an image:
This image is a simple black-to-transparent gradient saved in full RGBA PNG.
Here is the same image, converted to indexed-alpha PNG by GIMP (Photoshop produces the same result)
As you can see, the gradient is now half-opaque, half-transparent.
Here is the same image again, only this time it was converted to indexed-alpha PNG by a PHP script I wrote:
So my question is: Why are GIMP and Photoshop unable to support partial transparency in indexed images, when the PHP script clearly shows that such an image can be created with no problems?
Is there anything "wrong" with an image whose pallette contains alpha information?
A more programming-related question: Does this transparency in the last image work in Internet Explorer 6?
I've finally found the actual answer: There is a metadata entry that allows you to define the alpha value of each colour in the colour table. Most graphics programs don't make use of this, but it does exist and can be used, in particular by GD.
Another option besides fireworks is pngquant, a command line application that will convert a rgba png into an indexed png with transparency.
I found this post which talks some more about how to use it.
IE6 and earlier in windows does not support variable transparency PNGs without annoying workarounds. An indexed PNG will only show the fully opaque parts which usually works pretty well. A drop shadow would disappear but the opaque parts of the logo or icon would continue to show.
This page has a better explanation and instructions with more png compression and quantization tools: http://calendar.perfplanet.com/2010/png-that-works/
For the record, PNG does not literally support indexed images with an alpha channel. What is really happening is that PNG allows you to add additional colors to the color table (i.e. index) with alpha values in those colors... not a complete alpha channel. FWIW...
Yeah I know what you mean. Fireworks is the only image editing program that I know of that can create and edit PNG8+Alpha without problems. I wish more paint programs would support this format cause Fireworks is expensive!
I found a way in GIMP to create or convert an image with reduced color palette and alpha channel.
The trick is to add a mask to the layer.
Full steps to reproduce:
Have your image in one layer
Add a mask to the layer. Select Transfer layer's alpha channel.
Convert to Indexed (Image -> Mode -> Indexed...)
Save as PNG
Now your image has reduced colors and reduced size, but it keeps your smooth transparency.