THE MAGAZINE FOR COMPUTER APPLICATIONS
Circuit Cellar Online offers articles illustrating creative solutions
and unique applications through complete projects, practical
tutorials, and useful design techniques.

IMAGE PROGRESSING FUNDAMENTALS

Part 1: Playing with the Pixels

by James Antonakos

Start ý Imaging 101 ý RLE Compressor Tool ý Adjusting the X-Y Resolution ý Adjusting the Z-Axis Resolution ý Simple Image Operations ý Run-Length Compression ý More to Come ý Sources and PDF

RUN-LENGTH COMPRESSION

One of the simplest compression schemes available for images is Run-Length Encoding (RLE). In this technique, a single pixel value and a repeat count replace consecutive pixels in a single row of the image that have the same value. Figure 4 shows a simple example.

Figure 4ýThe format of the compressed RLE data is <repeat count>,<pixel value>. In this example, 20 bytes of pixel data are reduced to 12 bytes of RLE data, a compression of almost 50%.

The more the pixels stay the same, the greater the compression. For example, if the entire 256-pixel row of an image contains black pixels, the RLE data will consist of only two bytes (repeat count and pixel value). Photo 8 shows the results of RLE compressing the source image with the RLE Image Compressor.

Photo 8ýNotice that each line of pixels compresses differently.

Look at how well the rows at the top of the image have compressed compared to many of the rows making up the face and shirt. The long runs of black pixels in the top rows compress significantly. The rows of pixels making up the face and shirt have many intensity variations and thus do not compress as well.

To RLE compress the source image, left-click the Compress button. The processed image will be generated and the compression ratio computed. The ratio for the processed image in Photo 8 is 0.63, meaning the compressed image is 63% of its original size.

Select five or fewer bits per pixel, left-click Adjust Z to display the new processed image, and then click Compress to see the compression results. You will see that the compression ratio increases dramatically as fewer bits are used to represent a pixel. Table 2 shows all the compression results for the same source image. Note the last entry in the table. When using a single bit per pixel, the RLE compressed image only requires 3% of the original storage space. This is a nice feature of binary images.

Expanding the RLE data into the original image data is a simple matter of reading each <repeat count>,<pixel value> pair and repeating the pixel the appropriate number of times.

RLE compression is lossless, meaning that the expanded image data is exactly the same as the original image. RLE compression is used in the PCX image file format. There are many other compression methods, such as Huffman coding, LZW, and DCT. High-quality, color JPG image files are compressed using DCT, and the subsequent quantization of the coefficients makes it a lossy compression method. The uncompressed image is slightly different than the original, with little loss in quality. Lossy compression generally results in better compression than lossless compression.

PREVIOUS • NEXT