Image Recognition Dictionary Data Sharing for Mirror Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image recognizing apparatuses require increased dictionary data storage capacity due to the use of different dictionary data for similar image blocks, such as mirrored blocks, leading to inefficiencies in data management.
Innovation Solution
An image recognizing apparatus that determines whether a target block is a shared or mirror block based on its position, allowing common dictionary data to be used for both, thereby reducing the data amount required for image recognition and optimizing dictionary memory capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If different dictionary data is used for each block even when images are similar, then image recognition accuracy is maintained, but the data amount of dictionary data increases in proportion to the number of blocks
Solution Approach 1:
The patent merges dictionary data for mirror blocks by determining that blocks symmetric with respect to the vertical center line are mirror blocks, and using the same dictionary data for both the block and its mirror block. This combining approach reduces the total dictionary data quantity while maintaining recognition accuracy through symmetry exploitation.
Solution Approach 2:
The patent makes dictionary data universal by enabling a single dictionary data set to serve multiple purposes - it is used for both the original block and its mirror block. This multi-functionality allows the same dictionary data to recognize features in both symmetric blocks, reducing overall data requirements.
2Measurement precision
If different dictionary data is used for each block, then recognition of block-specific features is improved, but the capacity of the dictionary memory increases
Solution Approach 1:
The patent merges dictionary data for mirror blocks by determining that blocks symmetric with respect to the vertical center line are mirror blocks, and using the same dictionary data for both the block and its mirror block. This combining approach reduces the total dictionary data quantity while maintaining recognition accuracy through symmetry exploitation.
Solution Approach 2:
The patent applies local quality by treating different blocks differently based on their positional relationship - blocks are identified as either original blocks or mirror blocks, and dictionary data is selectively shared or differentiated accordingly. This localized approach optimizes memory usage while preserving necessary recognition capabilities.
3Quantity of substance
If dictionary data is shared between shared blocks and mirror blocks, then the data amount is reduced, but the system complexity increases due to block determination requirements
Solution Approach 1:
The patent segments blocks into three categories using a determination table: original blocks, mirror blocks, and shared blocks. This segmentation provides a systematic framework for managing dictionary data sharing, where the block determining module uses simple positional relationships (symmetry about the vertical center line) to classify blocks and select appropriate dictionary data.
Data Source
AI summary
An image recognizing apparatus includes a dictionary memory, a block determining module and a recognizing module. The dictionary memory stores dictionary data. The block determining module determines that a target block comprising a target pixel to be processed of a plurality of pixels in image data is a shared block to which the dictionary data is used or a mirror block to which the dictionary data to the shared block is used, based on a position of the target block. The recognizing module uses common dictionary data for the shared block and the mirror block, and recognizes a characteristic portion of the image expressed by the image data.