/ blog | image gallery | resize images online or on your PC maintaining quality

Direct links to popularity ranked list (best rated sites):

1. deviantart.com
2. mattosai.net
3. peorth.com.ar
4. animepaper.net
5. imageshack.us
6. saiyaman.info
7. desktopnexus.com

If you find an image that doesn’t fit your screen well, you can resize it here. It won’t degrade the quality of the wallpaper as other methods do (including Photoshop). You need a free account first. It’s very fast to create, and after that, you can use the quick add feature. You will just need to push a button, when viewing an image by itself in the browser. Or of course you can add the image manually.

There are a few situations when resizing images on reshade.com won’t give the expected results. Here’s why:

Compression artifacts

First, very small images are also highly compressed (jpeg for example). The compression leaves artifacts in the image. They appear like out of place rectangles mostly around edges. These are mostly poorly visible in a small image. They can be taken as background noise or possibly even over-sharpening side-effects. When enlarged they become a lot more apparent. Reshading reduces their influence most of the time. But, there are of course plenty of very strongly compressed images out there. This implies lots of artifacts. If also using strong sharpening when reshading they become more apparent than usual.

Noise-like details

Small images are usually created from some larger versions of photos or other works. By shrinking an image more detail is gathered in a tighter space. This means that many noise-like effects will be created. Reshade tries to make sense of edges only above a threshold, considering the rest as noise. And noise is not a good thing, so reshading removes it. Because of this, enlarging very small images sometimes doesn’t work perfectly.

Sub-optimal sharper value

In most situations, a value of 100(%) for the sharper setting when resizing will be the best choice. Some images may require less of more depending on the situation. Try not to use too large a value. Sometimes this will create very unpleasant effects. If this happens reduce the sharper value or un-check the corresponding box (this will set sharpening to 0%)

Big blurry originals

Reshading will work only for up-sizing images. When down-sizing another method is used. This will not de-blur the image as it is currently implemented. If you would like reshade.com to use deblurring on the original please tell me.

Upsize limits

Reshade.com allows only resizes up to approximately a maximum of 1920×1200 pixels. But, because cropping is enabled the actual zoom factor spans from 0% to 300%. There is a limit to the number of crops/day (about 50).

To begin with, various aspects about image resizing will be explained: what the problems are, how it works, who should be interested. To continue, a comparison will be made between existing methods. Finally, the conclusion will mention the best method for photo enlargement to use.

Problems to be solved by upsizing algorithms

One of them is edge pixel aliasing in the original images. This becomes even more apparent when enlarging. The edges in the enlarged image appear very jagged. With upsizing as with other algorithms there are many trade-offs. In one extreme you have edges that are sharp but jagged in another smooth but blurry. A compromise between smoothing an edge and keeping it sharp is searched for. But, over-smoothing and maintaining sharpness creates an unrealistic look to images. After enlargement photos start looking like paintings, and that makes it unacceptable sometimes. Another problem is loss of detailed texture in the upsized image. The only method that even considers this is reshading.

How enlarging works

Upsizing images increases their resolution (the size in pixels/points). Information from the original image is used to create an enlarged version. How to do this? Imagine a black and white image as a grid of pixels. They are spread out in a plane each having their own values. These values represent the brightness levels. In the original image the pixels are all side by side. Enlarging the image means expanding this grid. The more you stretch it, the more separated the pixels become. Those pixel places in-between need to be filled with their own values. This is called interpolation. There can be other processing steps after/before interpolation.

When to upsize an image

Increasing the resolution by 200% means transforming a 3 megapixel (MP) image into a 12 MP one. Doing this without losing quality is like transforming an average camera into a professional one. Cropping and enlarging a part of an picture is like having a better zoom lens. Also, starting with a professional camera and using upsizing would make very large prints easy.

Methods for enlarging pictures

Some well known approaches are commercially available: Photoshop, Genuine fractals, Photo zoom. There’s also a newer method that is free: Reshade.

Photoshop’s disability

Although everybody takes Photoshop as the tool for any image processing, its features cover only basic needs. It is designed to be a framework for image manipulation by allowing the inclusion of plug-ins that can easily interoperate. Even experienced Photoshop users have a hard time when enlarging pictures. The default methods for upsizing are made to be fast but not very powerful.

The other methods presented next do a better job at upsizing pictures. They are designed to allow greater enlargement factors without serious degradation in image quality.

Reshade

This is the newest among the ones mentioned. It can increase the size up to 200% without losing quality. Beyond this zoom, the pictures look a little “painted”. Even at higher zooms reshade works better on most images than any of the other techniques. In most cases a 200% upsize is the limit, because after that the sparseness of the details becomes obvious. Larger zooms can be useful in other situation. Large posters that are generally viewed from afar don’t need perfect quality when you look up-close. Try the free online image resizer implementing this method here.

Genuine fractals

This method tries not to overdo the sharpening and deblurring like photo zoom. For zoom factors of about 500% it maintains the quality of the image pretty well. For smaller zooms the image quality is very similar to what you can get in Photoshop. So, there still are blurry edges.

Photo zoom

Even though the clarity of the edges is better than for Genuine fractals sometimes (at large zoom factors, 400%), there is an obvious smoothing effect that creates the impression of over-painting. For photography, it adds an unrealistic look to the resized images. Although small details when enlarged remain smooth, reshading works better, also maintaining their aesthetics.

Conclusion

It is not suggested to use upsizing for every photo. The original can always look better even if an enlarged version has minor quality loss.

It often happens that you need to crop a part of a picture and it becomes too small. Or maybe you don’t have your best camera nearby and use a lower megapixel one. In these situations enlarging becomes the best alternative. I recommend reshading your images for best quality. You can try the online image resizing engine or get the desktop photo enlarger.