Image Capturing Device Characteristic Simplification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image capturing devices with object detection functions require significant memory and computing resources, leading to increased costs and reduced competitiveness, especially in compact handheld devices.
Innovation Solution
An image capturing device with a characteristic conversion module that simplifies characteristic value sets by using a lookup table to delete unnecessary value sets and replace them with matching sets, reducing memory requirements without affecting detection accuracy, and utilizing templates established through boosting, SVM, or PCA+LDA for object recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If object detection function is integrated into compact handheld camera device, then detection capability is improved, but memory requirement increases significantly
Solution Approach 1:
The patent transforms the full characteristic value sets into simplified characteristic value sets by changing the parameter representation. This is achieved through eigenface analysis where original image data is transformed into a reduced set of eigenvalues and eigenvectors that capture the essential features, thereby reducing memory requirements while maintaining detection capability
Solution Approach 2:
The patent extracts only the essential characteristic features from complete image data. By using eigenface decomposition, it separates and retains only the most significant components (eigenfaces with highest eigenvalues) that contribute to object recognition, discarding redundant information that would consume memory
2Adaptability or versatility
If object detection function is integrated into compact handheld camera device, then detection capability is improved, but computing resource consumption increases
Solution Approach 1:
The patent performs preliminary processing by pre-computing eigenfaces and eigenvalues from training images before actual detection. This preprocessing step creates a compact representation that can be quickly compared against new images, significantly reducing the computational burden during real-time detection operations in handheld devices
Solution Approach 2:
The patent changes the computational parameters from processing complete high-dimensional images to processing reduced-dimensional eigenvalue representations. This transformation reduces the complexity of comparison operations during detection, lowering computing resource consumption while preserving detection accuracy
3Measurement precision
If full characteristic value sets are stored, then detection accuracy is maintained, but memory cost increases
Solution Approach 1:
The patent changes the storage parameter from full-resolution characteristic value sets to compressed eigenvalue representations. By storing only the essential eigenfaces and their corresponding eigenvalues, it maintains the ability to accurately represent and recognize objects while dramatically reducing the memory space required
Solution Approach 2:
The patent creates simplified copies (eigenface representations) of original image characteristics that retain the essential recognition information. These compressed representations serve as efficient substitutes for storing and processing complete high-resolution characteristic data
Data Source
AI summary
An image capturing device capable of simplifying characteristic value sets of captured images and a control method thereof. The image capturing device comprises a characteristic conversion module, a data storage module, a characteristic simplification module, a template storage module and a recognition module. The characteristic conversion module converts an image captured by the image capturing device into a characteristic image, and the characteristic image includes a group of first characteristic value sets. The data storage module stores a lookup table which comprises second characteristic value sets. The characteristic simplification module performs a simplification process according to the lookup table to produce a simplified group of characteristic value sets. Finally, compares the simplified group of characteristic value sets with the plurality of templates stored in the template storage module to recognize a specific object in the image.