Inspection Algorithm for Hot Spot Detection in Plane Light Sources
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for inspecting luminance unevenness in edge-lit type plane light source devices, such as hot spots, are inefficient due to the need for multiple filters and are dependent on specific brightness-darkness patterns, making them time-consuming and inaccurate for various models with different light source arrangements.
Innovation Solution
A new evaluation value is calculated from a photographed image using an algorithm that assesses the difference between adjacent luminance values, allowing for automatic and objective detection of hot spots, independent of light source pitch and arrangement, using the expression HS = abs(val2 - val1) + abs(val2 - val3) / (val1 + val2 + val3), and determining the presence of hot spots based on threshold distances between extreme values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple filters are prepared to match different brightness-darkness patterns, then detection accuracy for various models is improved, but device complexity and processing time increase
Solution Approach 1:
The patent applies universality by creating a single evaluation algorithm that can detect hot spots across all plane light source device models regardless of their specific light source arrangements. Instead of preparing multiple model-specific filters, the invention uses a universal evaluation expression (HS = abs(val2 - val1) + abs(val2 - val3)) that adapts to different brightness-darkness patterns, light source pitches, and arrangements, making one algorithm serve multiple functions for all device variants.
Solution Approach 2:
The patent applies parameter changes by transforming the detection approach from using multiple fixed filters to using a single evaluation algorithm with dynamic parameters. The algorithm calculates evaluation values based on luminance values at different positions (val1, val2, val3) and uses threshold comparisons to detect hot spots. This allows the system to adapt to different device models by changing the input parameters (luminance values at different positions) rather than changing the fundamental detection method.
2Reliability
If scanning with multiple filters is performed to cover all brightness-darkness patterns, then comprehensive detection is improved, but processing time is prolonged
Solution Approach 1:
The patent applies segmentation by dividing the light-emitting surface into multiple measurement positions and calculating evaluation values for each position independently. The inspection region is divided into first, second, and third positions corresponding to different light source regions, allowing parallel calculation of evaluation values. This segmentation enables comprehensive detection across the entire surface while maintaining efficient processing through independent, parallel evaluations rather than sequential filter scanning.
Solution Approach 2:
The patent replaces the mechanical filter scanning system with a computational evaluation algorithm. Instead of physically scanning multiple filters across the device, the invention uses image capture to obtain luminance data and applies mathematical evaluation expressions to calculate hot spot detection values. This substitution of mechanical scanning with computational processing dramatically reduces processing time while maintaining comprehensive detection coverage.
3Adaptability or versatility
If visual sensory inspection is used, then flexibility in handling various patterns is maintained, but automation and objectivity are reduced
Solution Approach 1:
The patent applies self-service by enabling the inspection system to automatically adapt to different device models and hot spot patterns without human intervention. The evaluation algorithm automatically calculates evaluation values based on captured images and compares them against thresholds to determine hot spot presence. The system serves itself by autonomously handling various brightness-darkness patterns through the universal evaluation expression, eliminating the need for manual filter selection or adjustment while maintaining high adaptability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An inspecting device includes: an image acquiring unit which acquires a light-emitting surface image as a photographed image of the light-emitting surface; an inspecting unit which sets an inspecting range in a position in the light-emitting surface image in which the failure may appear, generates a one-dimensional luminance profile representing change in a luminance value along the first side in the inspecting range, detects an extreme value in the luminance profile, calculates an evaluation value for evaluating the difference between adjacent extreme values, and determines the presence or absence of the failure on the basis of the evaluation value; and an output unit which output information obtained by the inspecting unit.