Last updated on 5 Jan. 2005

Rambler's Top100

The commercial project made for presentation of new algorithm of scaling of raster images. All rights on application of the given algorithm belong to the author of the project.

Jpeg Repair RUS ENG








Smart Edges


Smart Picsel


New Mosaic








DJ-M-Spline - the new, linear and best quality algorithm of image cleaning.

JPEG compression breaks an image into rectangular blocks, each representing 8 x 8 pixels of detail, and then independently compresses each block. The result of JPEG compression on a file that was of poor original quality is an ugly, repetitive pattern of blocks. Sometimes the distortions are striking enough to be visible, and those are called artifacts.


There are some versions of artifacts:

Blocking artifacts:
Baseline JPEG can only work with 8x8 blocks, one at a time. This causes the best known artifact in compressed images - small square blocks all over the images, known both in JPEG and in video decoders.

Color distortion:
As human eyes are not as sensitive to color as to brightness, much of the detailed color (chrominance) information is disposed, while luminance is retained. This process is called "chroma subsampling", and it means that a color image is split into a brightness image and two color images.

Ringing artifacts:
Both JPEG2000 and JPEG operate in spectral domain, trying to represent the image as a sum of smooth oscillating waves.

Blurring artifacts:
Blurring means that the image is smoother than originally. The pictures were stored at approximately 2.0 bits per pixel, four times weaker compression rate as above.

Recently, on the market of professional graphic programmes, quite a few more effective and better quality algorithms have appeared. Along with the linear methods non linear (adaptive) algorithms appeared Net Image 5.1, Define, Cleanerzoomer, Jpeg Enhancer, Un Jpeg etc. Notwithstanding the diversity of existing methods of noise and removes of jpeg artifacts, Jpeg Enhancer and Cleanerzoomer are generally recognized as ones of the best quality and the most accurate. However, because none of the said methods does not get an identically good result, the work to create new algorithms is still ongoing with various success.


The aim of data spelling appears as the presentation of the new linear method which allows to carry out removes of jpeg artifacts a better quality, better than all known methods, both linear and adaptive. The main advantage of suggested method over the rest of them is the speed (because it appears linear) simpleness of its realization and best quality closest to achievable limits. The method brought to your attention was called Magic spline or M-spline by its author.

DJ-M-spline - algorithm is fully worked out and realized exclusively by the author and all the rights of its use until the decision of sale, belong to the author.

DJ-M-spline2 algorithm is a newer version of DJ-M-spline. New updating of algorithm which differs from the previous version following technological decisions.


Smarp Pixel - Adaptation of a spline to local features of scaled area.

Smarp Function -Introduction in function of a spline of nonlinear operators.

Smart Edge - Adaptation of a spline to edges of scaled area.


Therefore the mistake of JPEG cleaning is reduced, having reached new record values.

In order to confirm above, I suggest to get familiar with practical realization of the algorithm on the well known examples of tasks and solutions.

As the test the following representation has been chosen:

flowers lhouse2

To see the picture click on the fragment

Test flowers

The compressed image has been restored by means of algorithm DJ-spline2.

For comparison show how to work one of the best known methods of removal JPEG artifacts - Cleanerzoomer.

The resolution of image will be enlarged by 4 times using a box convolution kemel.


Original Compressed Cleanerzoomer DJ-M-Spline2

(New method)

RMSE 7.10 RMSE 7.00 RMSE 6.77

Test flowers

Chrominance component

The resolution of image will be enlarged by 4 times using a box convolution kemel.


Compressed Cleanerzoomer DJ-M-Spline2

(New method)
Often happens it is necessary to increase the image size include by JPEG artifacts.
The algorithm DJ-M-Spline can simultaneously both removal artifacts and increase size the image.
Practically the algorithm DJ-M-Spline is unique qualitative way of the solve of such task.

Test - lhouse

Original image was reduced 0.5X using a box convolution kemel and has been compressed.

The compressed image has been restored by means of algorithm DJ-spline2 and simultaneously with it

will be enlarged by 2 times.

For comparison show how to work one of the best known methods of removal JPEG artifacts - Cleanerzoomer.

After removal artifacts the received image has been increased by one of the most widespread methods - Bicubic.


Original Reduced & Compressed Cleanerzoomer DJ-M-Spline2

enlarged by 2 times (bicubic method) (New method)

RMSE 12.73 RMSE 14.29 RMSE 10.92
The real images.
It is photographed by the SONY DSC-U10

Real image DJ-M-Spline2

(New method)
Your comments and questions:


Hosted by uCoz