Image Encoding via Selective Run-Length and Gradient Compression

The patent segments the image into multiple blocks and processes each block independently using different encoding schemes (RLE, gradient encoding, or DCT). This segmentation allows the system to apply computationally simple methods to uniform regions while using more complex methods only where necessary, thereby achieving good compression ratios without uniformly high power consumption across the entire image.

The patent applies different encoding qualities to different regions of the image based on their characteristics. Uniform regions use simple RLE encoding, smoothly varying regions use gradient encoding, and complex regions use DCT. This local adaptation of encoding quality ensures that power consumption is optimized for each region's specific needs, achieving overall compression without unnecessary computational overhead in simple regions.

The patent dynamically selects the encoding scheme for each block based on its characteristics. The system evaluates each block and adapts the encoding method (RLE, gradient encoding, or DCT) to match the block's content. This dynamic adaptation reduces overall computational complexity by avoiding the application of complex DCT to simple uniform blocks, while still achieving high compression ratios where needed.

The patent changes the encoding parameter (encoding scheme) based on the block's characteristics. By monitoring block uniformity and variation, the system switches between different encoding parameters (RLE for uniform, gradient for smooth, DCT for complex), thereby managing device complexity adaptively while maintaining compression effectiveness.

The patent applies different levels of compression effort to different regions based on their characteristics. Uniform regions use lightweight RLE encoding with low power consumption, while complex regions use more aggressive DCT-based compression. This local differentiation allows the system to achieve high overall compression efficiency without uniformly high power consumption, as simple regions don't waste energy on unnecessary compression.

The patent applies compression selectively rather than uniformly. By using simple RLE for uniform regions and reserving complex DCT for only those regions that benefit from it, the system avoids excessive computational action in regions where it isn't needed, thereby reducing power consumption while maintaining adequate compression efficiency for the overall image.