Image Analysis Apparatus Using Segmented Down-Sampling for Focus State Determination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional surveillance cameras face challenges in maintaining focus due to weather conditions or fatigue, leading to blurry images, and existing image analysis methods require large memory and complex computations to determine focus state, resulting in slow classification results.
Innovation Solution
An image analysis method and apparatus that divides original images into auxiliary images using down-sampling technology, applying these images to an image analysis model to rapidly and accurately determine focus state without retraining the base model, allowing for dynamic adjustment of classification criteria.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the surveillance camera analyzes spatial domain information of the captured image to determine focus state, then the focus state can be determined, but large-capacity memory is required to store spatial domain information and complex computation process results in lengthy computation time
Solution Approach 1:
The patent divides the original image into multiple sub-images (auxiliary images) and processes them separately. Instead of analyzing the entire large image at once, the system segments it into manageable portions, processes each portion independently to determine local focus states, and then integrates these results. This segmentation approach reduces the computational burden on memory and processing units while maintaining overall focus determination accuracy.
2Device complexity
If the captured image is divided into several auxiliary images for analysis, then the computation process can be simplified, but the classification result is still determined by long-term computation
Solution Approach 1:
The patent extracts and utilizes down-sampled versions of the image (reduced-resolution auxiliary images) to perform initial focus state analysis. By taking out a down-sampled representation that contains essential focus information but occupies less computational resources, the system can quickly determine focus states without processing the full-resolution image, thereby reducing computation time while maintaining adequate accuracy.
3Productivity
If down-sampling technology is applied to divide the original image into auxiliary images, then computational load and memory requirements are reduced, but image classification accuracy may be compromised
Solution Approach 1:
The patent employs a dynamic multi-scale analysis approach where images are processed at different resolution levels. The system dynamically adjusts the analysis strategy by first performing quick assessment on down-sampled auxiliary images to identify regions of interest, then selectively applying more detailed analysis only to those specific regions at higher resolutions. This dynamic approach maintains high classification accuracy while minimizing overall computational load.
Data Source
AI summary
An image analysis method is applied to an image analysis apparatus having an operation processor and an image receiver. The image analysis method includes setting a range provided by an original image as a reference image, and dividing the reference image into a plurality of first auxiliary images in accordance with a valid size. The plurality of first auxiliary images is applied for an image analysis model to generate an image classification result. The operation processor divides the base image acquired by the image receiver into a plurality of second auxiliary images in accordance with the valid size, and the plurality of second auxiliary images is applied for the image analysis model to decide a number of the plurality of first auxiliary images.