Image Attribute Discrimination Excluding Heterogeneous Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image attribute discrimination techniques fail to accurately discriminate image attributes when heterogeneous regions, such as added objects or shadows, are present in the image data, leading to false or low-likelihood scene discrimination.
Innovation Solution
An image attribute discrimination apparatus that specifies and excludes heterogeneous regions from the feature extraction process, using a heterogeneous region specifying unit to identify and isolate these areas, and an attribute discrimination unit to discriminate based on features from the remaining regions, thereby improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature quantity is extracted from the whole image data including heterogeneous regions, then the processing is simple and fast, but the attribute discrimination accuracy deteriorates due to false scene discrimination
Solution Approach 1:
The image data is segmented into heterogeneous regions and non-heterogeneous regions. The heterogeneous region specifying unit identifies and separates these regions, allowing feature quantity extraction to be performed only on the non-heterogeneous regions, thereby improving attribute discrimination accuracy without excessively increasing processing complexity
Solution Approach 2:
The heterogeneous regions are extracted and excluded from the feature quantity extraction process. By removing these problematic regions from consideration, the system avoids false scene discrimination while maintaining relatively simple processing through targeted feature extraction from only the relevant non-heterogeneous regions
2Measurement precision
If heterogeneous regions are excluded from feature extraction, then attribute discrimination accuracy improves, but processing time increases due to additional region specification steps
Solution Approach 1:
The heterogeneous region specifying unit performs preliminary identification of heterogeneous regions before feature quantity extraction. By pre-marking these regions for exclusion, the subsequent feature extraction process can efficiently skip them without repeated analysis, reducing the overall time penalty of the additional specification step
3Reliability
If conventional feature extraction is used without heterogeneous region detection, then processing is efficient, but false scene discrimination occurs leading to incorrect attribute identification
Solution Approach 1:
The system segments the image into heterogeneous and non-heterogeneous regions, enabling reliable feature extraction only from appropriate areas. This segmentation approach maintains system complexity at an acceptable level while dramatically improving scene discrimination reliability by excluding contaminating heterogeneous regions
Solution Approach 2:
The heterogeneous region specifying unit acts as an intermediary between the raw image data and the feature extraction process. It filters out problematic heterogeneous regions before they can corrupt the scene discrimination, thereby improving reliability without requiring complete redesign of the overall system architecture
Data Source
AI summary
An attribute of image data can accurately be discriminated. An image attribute discrimination apparatus includes a heterogeneous region extracting unit that specifies a heterogeneous region from image data. The heterogeneous region includes a heterogeneous matter whose attribute is different from that of a content originally produced by the image data. An image attribute discrimination apparatus further includes a scene discrimination unit that discriminates the attribute of the image data based on a feature quantity extracted from a pixel group except each pixel in the heterogeneous region in each pixel of the image data.