I'm not very hopeful that this is actually even possible because JPG is a lossy compression format, but I'll ask anyway.
I have some incomplete metadata about an image, and need to figure out the dimensions without loading the file into memory, or using command-line tools like imagemagick.
This is what I know:
file format is JPG (JPG is an 8-bit image format).
total file size is 96284 bytes.
image width is 600 pixels.
Assuming the height could be variable, is there a mathematical equation I can use to calculate the height predictably?
No, there is no relation between the number of pixels and the compressed size.
You can try your luck by analyzing the file header to get the width/height tags. A library like jpeglib allows you to read just the header, but it takes some effort to integrate it.
Ok, so I tried to use the imagemagick command:
"convert picA.png -interlace line picB.png"
to make an interlace version of my .png images. Most of the time, I got the resulting image is larger than the original one, which is kinda normal. However, on certain image, the resulting image size is smaller.
So I just wonder why does that happen? I really don't want my new image to lose any quality because of the command.
Also, is there any compatibility problem with interlaced .png image?
EDIT: I guess my problem is that the original image was not compressed as best as it could be.
The following only applies to the cases where the pixel size is >= 8 bits. I didn't investigate for other cases but I expect similar outcomes.
A content-identical interlaced PNG image file will almost always be greater because of the additional data for filter type descriptions required to handle the passes scanlines. This is what I explained in details in this web page based on the PNG RFC RFC2083.
In short, this is because the sum of the below number of bytes for interlaced filter types description per interlacing pass is almost always greater than the image height (which is the number of filter types for non-interlaced images):
nb_pass1_lines = CEIL(height/8)
nb_pass2_lines = (width>4?CEIL(height/8):0)
nb_pass3_lines = CEIL((height-4)/8)
nb_pass4_lines = (width>2?CEIL(height/4):0)
nb_pass5_lines = CEIL((height-2)/4)
nb_pass6_lines = (width>1?CEIL(height/2):0)
nb_pass7_lines = FLOOR(height/2)
Though, theoretically, it can be that the data entropy/complexity accidentally gets lowered enough by the Adam7 interlacing so that, with the help of filtering, the usually additional space needed for filter types with interlacing may be compensated through the deflate compression used for the PNG format. This would be a particular case to be proven as the entropy/complexity is more likely to increase with interlacing because the image data is made less consistent through the interlacing deconstruction.
I used the word "accidentally" because reducing the data entropy/complexity is not the purpose of the Adam7 interlacing. Its purpose is to allow the progressive loading and display of the image through a passes mechanism. While, reducing the entropy/complexity is the purpose of the filtering for PNG.
I used the word "usually" because, as shown in the explanation web page, for example, a 1 pixel image will be described through the same length of uncompressed data whether interlaced or not. So, in this case, no additional space should be needed.
When it comes to the PNG file size, a lower size for interlaced can be due to:
Different non-pixel encoding related content embedded in the file such as palette (in the case of color type =! 3) and non-critical chunks such as chromaticities, gamma, number of significant bits, default background color, histogram, transparency, physical pixel dimensions, time, text, compressed text. Note that some of those non-pixel encoding related content can lead to different display of the image depending on the software used and the situation.
Different pixel encoding related content (which can change the image quality) such as bit depth, color type (and thus the use of palette or not with color type = 3), image size,... .
Different compression related content such as better filtering choices, accidental lower data entropy/complexity due to interlacing as explained above (theoretical particular case), higher compression level (as you mentioned)
If I had to check whether 2 PNG image files are equivalent pixel wise, I would use the following command in a bash prompt:
diff <( convert non-interlaced.png rgba:- ) <( convert interlaced.png rgba:- )
It should return no difference.
For the compatibility question, if the PNG encoder and PNG decoder implement the mandatory aspects of the PNG RFC, I see no reason for the interlacing to lead to a compatibility issue.
Edit 2018 11 13:
Some experiments based on auto evolved distributed genetic algorithms with niche mechanism (hosted on https://en.oga.jod.li ) are explained here:
https://jod.li/2018/11/13/can-an-interlaced-png-image-be-smaller-than-the-equivalent-non-interlaced-image/
Those experiments show that it is possible for equivalent PNG images to have a smaller size interlaced than non-interlaced. The best images for this are tall, they have a one pixel width and have pixel content that appear random. Though, the shape is not the only important aspect for the interlaced image to be smaller than the non-interlaced image as random cases with the same shape lead to different size differences.
So, yes, some PNG images can be identical pixel wise and for non-pixel related content but have a smaller size interlaced than non-interlaced.
So I just wonder why does that happen?
From section Interlacing and pass extraction of the PNG spec.
Scanlines that do not completely fill an integral number of bytes are padded as defined in 7.2: Scanlines.
NOTE If the reference image contains fewer than five columns or fewer than five rows, some passes will be empty.
I would assume the behavior your experiencing is the result of the Adam7 method requiring additional padding.
I would like to know why we need to decode let's say a png to a bitmap in order to show the image.
Why not just show the png like that (encoded).
I'm asking here a moron type of question on purpose. It's clear to me it's impossible to show an encoded image just like that but I want to know why, and how an image is shown on a screen because it's easy just to do :
canvas.drawBitmap(((AndroidImage)Image).bitmap, x, y, null);
I want to understand the full of it. I'm guessing we need to show every pixels one by one, but I want more details.
It's easy to know how to do, it's a bit harder to understand why.
If someone has a course/tuto/article/explanation that explains it... I would appreciate
Thanks in advance
PS : Please don't respond "you need to decode/convert png to bitmap" I know that... And that's not my question
There are lots of reasons. There is not really a direct relation between 'a value in a file' and 'a pixel on a screen'.
You need to know the width and height of the bitmap. You cannot infer this from the image size -- it has to be stored somewhere inside the image file. (Or anywhere else. Point is, you have to know its size.)
You need to know the bit depth and color model of the bitmap. You cannot meaningfully copy an 8-bit indexed image directly onto a screen that accepts 32-bit BGR ordering with an unused byte, for example.
Your example, the PNG file format, specifies that all image data is compressed. This is for a sane reason: the PNG format was designed for use on web pages, in a time period where every byte still counted. But even the lowly simple BMP file format uses a very specific form of 'encoding': in its 24-bit format, every line consists of sets of BGR values for each pixel and is padded at the end with enough bytes to make its total length evenly divisible by 4.
JPEG uses an even more advanced encoding scheme (which is too difficult to explain in a few short words) so it can compress images even more. The advanced encoding scheme allows far more compression than regular methods (which in turn means there is only the tiniest relation between 'values in the file' and 'pixels on the screen').
I have a series of images that change relatively little during the day, but suffer from severe over and underexposure sometimes. There's nothing I can do in the way of fixing the loss of details, however I want to make it so that one from to the next is not so jolting. In other words, I'd like to normalize brightness / color levels the sequence of images so that it feels more consistent from one from to the next. Ultimately this sequence is being encoded to a video and therefore why this is important to me.
How can I normalize these values for a series of a few hundred frames via ImageMagick or some other means?
Use ImageMagick command-line's -equalize option, that should do the trick. This will only adjust the current image to use its full color range properly.
If you want one image to match it's histogram and/or saturation to another, use this ImageMagick script.
Consider an application handling uploading of potentially very large PNG files.
All uploaded files must be stored to disk for later retrieval. However, the PNG files can be up to 30 MB in size, but disk storage limitations gives a maximum per file size of 1 MB.
The problem is to take an input PNG of file size up to 30 MB and produce an output PNG of file size below 1 MB.
This operation will obviously be lossy - and reduction in image quality, colors, etc is not a problem. However, one thing that must not be changed is the image dimension. Hence, an input file of dimension 800x600 must produce an output file of dimension 800x600.
The above requirements outlined above are strict and cannot be changed.
Using ImageMagick (or some other open source tool) how would you go about reducing the file size of input PNG-files of size ~30 MB to a maximum of 1 MB per file, without changing image dimensions?
PNG is not a lossy image format, so you would likely need to convert the image into another format-- most likely JPEG. JPEG has a settable "quality" factor-- you could simply keep reducing the quality factor until you got an image that was small enough. All of this can be done without changing the image resolution.
Obviously, depending on the image, the loss of visual quality may be substantial. JPEG does best for "true life" images, such as pictures from cameras. It does not do as well for logos, screen shots, or other images with "sharp" transitions from light to dark. (PNG, on the other hand, has the opposite behavior-- it's best for logos, etc.)
However, at 800x600, it likely will be very easy to get a JPEG down under 1MB. (I would be very surprised to see a 30MB file at those smallish dimensions.) In fact, even uncompressed, the image would only be around 1.4MB:
800 pixels * 600 pixels * 3 Bytes / color = 1,440,000 Bytes = 1.4MB
Therefore, you only need a 1.4:1 compression ratio to get the image down to 1MB. Depending on the type of image, the PNG compression may very well provide that level of compression. If not, JPEG almost certainly could-- JPEG compression ratios on the order of 10:1 are not uncommon. Again, the quality / size of the output will depend on the type of image.
Finally, while I have not used ImageMagick in a little while, I'm almost certain there are options to re-compress an image using a specific quality factor. Read through the docs, and start experimenting!
EDIT: Looks like it should, indeed, be pretty easy with ImageMagick. From the docs:
$magick> convert input.png -quality 75 output.jpg
Just keep playing with the quality value until you get a suitable output.
Your example is troublesome because a 30MB image at 800x600 resolution is storing 500 bits per pixel. Clearly wildly unrealistic. Please give us real numbers.
Meanwhile, the "cheap and cheerful" approach I would try would be as follows: scale the image down by a factor of 6, then scale it back up by a factor of 6, then run it through PNG compression. If you get lucky, you'll reduce image size by a factor of 36. If you get unlucky the savings will be more like 6.
pngtopng big.png | pnmscale -reduce 6 | pnmscale 6 | pnmtopng > big.png
If that's not enough you can toss a ppmquant in the middle (on the small image) to reduce the number of colors. (The examples are netpbm/pbmplus, which I have always found easier to understand than ImageMagick.)
To know whether such a solution is reasonable, we have to know the true numbers of your problem.
Also, if you are really going to throw away the information permanently, you are almost certainly better off using JPEG compression, which is designed to lose information reasonably gracefully. Is there some reason JPEG is not appropriate for your application?
Since the size of an image file is directly related to the image dimensions and the number of colours, you seem to have only one choice: reduce the number of colours.
And ~30MB down to 1MB is a very large reduction.
It would be difficult to achieve this ratio with a conversion to monochrome.
It depends a lot on what you want at the end, I often like to reduce the number of colors while perserving the size. In many many cases the reduced colors does not matter. Here is an example of reducing the colors to 254.
convert -colors 254 in.png out.png
You can try the pngquant utility. It is very simple to install and to use. And it can compress your PNGs a lot without visible quality loss.
Once you install it try something like this:
pngquant yourfile.png
pngquant --quality=0-70 yourfile.png
For my demo image (generated by imagemagick) the first command reduces 350KB to 110KB, and the second one reduces it to 65KB.
Step 1: Decrease the image to 1/16 of its original size.
Step 2: Decrease the amount of colors.
Step 3: Increase the size of the image back to its original size.
I know you want to preserve the pixel size, but can you reduce the pixel size and adjust the DPI stored with the image so that the display size is preserved? It depends on what client you'll be using to view the images, but most should observe it. If you are using the images on the web, then you can just set the pixel size of the <img> tag.
It depends on they type of image, is it a real life picture or computer generated image,
for real life images png will do very little it might even not compress at all, use jpg for those images, it the image has a limited number of different colors (it can have a 24 bit image depth but the number of unique images will be low) png can compress quite nicely.
png is basicly an implementation of zip for images so if a lot of pixels are the same you can have a rather nice compression ratio, if you need lossless compression don't do resizing.
use optipng it reduce size without loss
http://optipng.sourceforge.net/
Try ImageOptim https://imageoptim.com/mac it is free and open source
If you want to modify the image size in ubuntu, you can try "gimp".
I have tried couple of image editing apps in ubuntu and this seemed to be the best among them.
Installation:
Open terminal
Type: sudo apt install gimp-plugin-registry
Give admin password. You'll need net connection for this.
Once installed, open the image with GIMP image editor. Then go to: File > Export as > Click on 'Export' button
You will get a small window, where check box on "Show preview in image window". Once you check this option, you will get to see the current size of the file along with Quality level.
Adjust the quality level to increase/decrease the file size.
Once adjusting is done, click on 'Export' button finally to save the file.
Right click on the image. Select open with paint. Click on resize. Click on pixel and change the horizontal to 250 or 200.
That's the only thing. It is the fastest way for those who are using Windows XP or Windows 7.