Feature Descriptor Encoding Using Differential Index Sorting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing feature descriptor encoding systems face issues with large data sizes due to fixed bit-length encoding of coordinate values, leading to increased communication time and reduced memory storage capacity for image matching and retrieval, particularly in large-scale image processing.
Innovation Solution
A feature descriptor encoding apparatus that includes a feature point detection unit, a local feature descriptor extracting unit, a coordinate value scanning unit for converting coordinates into index values, a sorting unit, a difference calculation unit, and a differential encoding unit that encodes differential values in a sequential permutation, reducing the size of encoded feature descriptors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If coordinate values are encoded with fixed bit length, then encoding simplicity is maintained, but data size increases significantly
Solution Approach 1:
The patent changes the encoding parameter from fixed bit length to variable bit length based on the actual coordinate values. By calculating the number of bits required to represent each coordinate value and using that as the encoding length, the system reduces data size while maintaining encoding simplicity through a straightforward calculation-based approach.
2Ease of manufacture
If fixed bit length encoding is used for coordinate values, then encoding process is simple, but communication time increases
Solution Approach 1:
The patent applies parameter changes by transitioning from fixed bit length to variable bit length encoding. The encoding length is dynamically determined based on the actual coordinate values, which reduces the total number of bits transmitted and consequently decreases communication time while keeping the encoding process simple through mathematical calculation.
3Ease of manufacture
If fixed bit length encoding is used, then implementation is straightforward, but memory storage capacity is reduced
Solution Approach 1:
The patent implements parameter changes by using variable bit length encoding that adapts to the actual coordinate values. This approach increases memory storage capacity by reducing the total data size required to store coordinate information, while maintaining implementation straightforwardness through a simple calculation-based encoding method.
Data Source
AI summary
An object of the present invention is to reduce the size of encoded feature descriptors. A feature descriptor encoding apparatus includes a feature point detection unit that detects feature points from an image and outputs coordinate values; a local feature descriptor extracting unit that extracts local feature descriptors from local regions centered on the coordinate values of the feature points; a coordinate value scanning unit that converts the coordinate values into index values according to a specific scanning method; a sorting unit that sorts the index values; a difference calculation unit that calculates a differential value between two adjacent index values of the sorted index values and outputs a sequence of differential values; a differential encoding unit that encodes the sequence of the differential values in a sequential permutation; and a local feature descriptor encoding unit that encodes the local feature descriptors of the corresponding feature points in the same permutation as that of the sorted index values.