Image Processing Keypoint Density Restriction and Reference Update
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image comparison technologies fail to accurately identify and trace target objects due to inappropriate keypoint selection and outdated reference images, especially when environmental conditions change, leading to unsuccessful object identification.
Innovation Solution
An image processing method that determines representative keypoints by a density restriction-based approach, distributing them uniformly across the image and updates reference images based on comparison results to adapt to changing conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional image comparison technologies select pixels with stronger feature values as keypoints, then the keypoint selection process is simple and fast, but the keypoints become excessively concentrated in areas with obvious features (e.g., non-target objects), causing target objects with unobvious features to be missed
Solution Approach 1:
The image is divided into multiple regions or areas, and keypoints are selected within each region independently. This segmentation approach prevents excessive concentration of keypoints in a single high-contrast area while ensuring coverage across the entire image, thereby improving target detection accuracy without significantly increasing computational complexity
Solution Approach 2:
Different regions of the image are treated differently in the keypoint selection process. Regions containing target objects (even with unobvious features) are given special consideration through localized keypoint selection strategies, while high-contrast non-target regions have their keypoint density controlled. This local differentiation ensures that keypoints are distributed more uniformly across the image
2Device complexity
If conventional image comparison technologies use fixed reference images, then the system structure is simple, but the system cannot adapt to changing environmental conditions (brightness, shooting angle, texture changes), leading to identification failure
Solution Approach 1:
The reference image system is made dynamic by automatically updating reference images based on newly detected target objects. Instead of using static, pre-stored reference images, the system continuously adapts its reference library by incorporating newly identified targets, allowing it to respond to environmental changes such as brightness variations, angle changes, and texture modifications
Solution Approach 2:
The system implements a feedback mechanism where detection results are fed back into the reference image database. When a target object is successfully identified, its image data is used to update or add new reference images, creating a self-improving system that adapts to changing conditions over time
Data Source
AI summary
An image processing apparatus and method thereof are provided. The image processing apparatus stores at least a reference image and performs the following operations: (a) receiving an image, (b) determining a plurality of representative keypoints for the image, such as determining the representative keypoints by a density restriction based method, (c) finding out that a matched area in the image corresponds to a first reference image according to the representative keypoints, (d) determining that a matched number between the representative keypoints and a plurality of reference keypoints of the first reference image is less than a threshold, and (e) storing the matched area in the image processing apparatus as a second reference image.