Multi-Exposure Image Inspection for Gradation Difference Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing inspecting apparatuses face challenges in accurately distinguishing between non-defective and defective products when the difference is small or when the material, shape, or photographing conditions of the inspection target object change, leading to difficulties in setting suitable exposure times for image recognition.
Innovation Solution
An inspecting apparatus that uses an image pickup unit to capture images with different exposure times, generates weighted image data by emphasizing regions with significant gradation differences, and combines these data to determine the inspection region's state using a reference image, employing filters like Laplacian or Gaussian filters to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single exposure time is used for image capture, then the device complexity is reduced, but the measurement precision deteriorates when the difference between non-defective and defective products is small or when material/shape/photographing conditions change
Solution Approach 1:
The patent segments the image capture process into multiple images with different exposure times (first exposure time for bright regions, second exposure time for dark regions). This allows each region to be captured under optimal exposure conditions, improving defect detection accuracy without requiring a single complex exposure time setting for the entire image.
Solution Approach 2:
The patent adds a new dimension to image capture by introducing multiple exposure times as an additional parameter. Instead of capturing a single image with one exposure time, the system captures multiple images varying the exposure time dimension, enabling better differentiation between defective and non-defective regions even when the difference is subtle.
2Measurement precision
If multiple images with different exposure times are captured and combined, then the measurement precision for defect detection is improved, but the device complexity and processing requirements increase
Solution Approach 1:
The patent applies local quality by selectively combining image regions based on their specific characteristics. Bright regions use the first exposure time image while dark regions use the second exposure time image. The system determines appropriate combination methods for different regions, optimizing defect detection accuracy for each local area without uniformly processing the entire image.
Solution Approach 2:
The patent introduces an intermediary processing stage that includes a determination unit and combination unit. These intermediary components analyze the multiple captured images and mediate the combination process by selecting appropriate images and regions based on predetermined conditions, simplifying the overall processing complexity while maintaining high measurement precision.
3Measurement precision
If the exposure time is optimized for bright regions, then the bright region detail is improved, but the dark region detail deteriorates. Conversely, if optimized for dark regions, then dark region detail is improved, but bright region detail deteriorates
Solution Approach 1:
The patent segments the image into bright regions and dark regions based on luminance thresholds. Different exposure time images are selectively applied to different segments: the first exposure time image for bright regions and the second exposure time image for dark regions. This segmentation allows each region to be optimized independently for its specific lighting conditions.
Solution Approach 2:
The patent implements dynamic adaptation by adjusting the exposure time based on the specific characteristics of each region. The system dynamically selects which exposure time image to use for each region rather than using a fixed exposure time for the entire image. This dynamic approach enhances the system's adaptability to varying material properties, shapes, and lighting conditions.
Data Source
AI summary
An inspecting apparatus includes an image pickup unit configured to pick up a plurality of images of an inspection target object with different exposure times and generate, image data of an inspection target object image including an inspection region, a weighted-image-data generating unit configured to weight, for each of the image data generated with the exposure times different from one another, data of pixels indicating a region where a difference in gradation of pixel values is relatively large among regions of pixels included in the image data and generate weighted image data, an image-data combining unit configured to generate combined image data obtained by combining the generated respective weighted image data, and a determining unit configured to determine a state of the inspection region on the basis of image data of a reference image set and the generated combined image data.