JPEG Decoder with Resize and Reorder Units for Memory Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
JPEG image decoding and viewing require large memory buffers, leading to increased costs and slower processing due to the need for significant memory bandwidth, especially when handling full-color images.
Innovation Solution
A system and method that includes a decoder, a resize unit, a memory, and a reorder data unit to decompress, resize, and rearrange image data for efficient display, reducing buffer size requirements and improving processing speed by resizing images along axes and rearranging data formats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If JPEG image decoding is performed before displaying the image, then the image can be decoded completely, but a large buffer is required (on the order of the largest supported image size)
Solution Approach 1:
The patent divides the image decoding and display process into segments that can be handled independently. Instead of decoding the entire image before display, the system decodes and displays image blocks sequentially. The reorder address generator creates segment-specific reorder addresses that allow each block to be processed and displayed independently, eliminating the need for a large buffer to hold the entire decoded image.
Solution Approach 2:
The patent performs preliminary reordering of image blocks based on display requirements before actual display. The reorder address generator pre-calculates the correct display order of blocks and prepares reorder addresses in advance, allowing the display system to efficiently render blocks in the correct sequence without requiring complete prior decoding of the entire image.
2Adaptability or versatility
If a large buffer is allocated to store JPEG image data, then maximum image size can be supported, but memory costs and processing time increase
Solution Approach 1:
The patent implements a dynamic buffer management system where the buffer size adapts to the actual display requirements rather than being fixed at the maximum supported image size. The reorder address generator dynamically calculates buffer addresses based on the current block being processed and its display position, allowing the system to support maximum image sizes while using only the minimal necessary buffer space at any given time.
Solution Approach 2:
The patent applies different buffer management strategies to different regions of the image based on local display requirements. Each image block is processed with its own specific reorder address calculation, allowing the buffer to be efficiently utilized for only the relevant portion of the image being currently decoded and displayed, rather than allocating space for the entire image.
3Reliability
If the entire JPEG image is decoded before viewing, then complete image data is available, but significant memory bandwidth is required
Solution Approach 1:
The patent extracts and processes only the necessary image blocks for current display rather than decoding the entire image. The reorder address generator identifies and extracts specific blocks that need to be displayed next, decoding and transferring only those blocks through the memory system. This extraction approach ensures complete image data can be reconstructed over time while dramatically reducing the memory bandwidth required at any moment.
Data Source
AI summary
System and method for memory efficient decode and viewing of compressed image files. A preferred embodiment comprises a decoder coupled to a data input, a resize unit coupled to the decoder, a memory coupled to the resize unit, and a reorder data unit coupled to the memory. The decoder decompresses image data that is provided by the data input, while the resize unit resizes the decompressed image data from a first size to a second size. The memory stores the resized image data and the reorder data unit rearranges the image data into a format suitable for display. By resizing the image data prior to subsequent processing, memory requirements (such as storage space and bandwidth) are reduced.