TOF Image Processing With Adaptive Exposure for Mixed Reflectivity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D visual imaging technologies using Time of Flight (TOF) sensors often result in overexposure or underexposure, particularly when objects with high and low reflectivity are present, leading to poor image quality due to data loss and underexposure of low reflectivity objects.
Innovation Solution
A method and system that adjusts exposure time by calculating the proportion of overexposed and underexposed pixel points, replacing overexposed pixels with short exposure time data to improve image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If automatic exposure reduces integration time to reduce overexposed points, then overexposure is reduced, but low reflectivity objects become underexposed in large areas causing data loss
Solution Approach 1:
The patent applies dynamic exposure time adjustment by calculating the proportion of overexposed and underexposed pixel points and adjusting the exposure time accordingly. The system dynamically switches between long exposure time (for capturing low reflectivity objects) and short exposure time (for avoiding overexposure of high reflectivity objects) based on real-time scene analysis, resolving the contradiction between preventing overexposure and avoiding data loss.
Solution Approach 2:
The patent segments the image data into different exposure time datasets (long exposure time data and short exposure time data) and selectively combines them. By segmenting the pixel points into overexposed, underexposed, and normal categories, the system can apply different exposure time data to different regions, ensuring that high reflectivity objects use short exposure data while low reflectivity objects use long exposure data.
2Loss of information
If long exposure time is used to capture low reflectivity objects, then low reflectivity objects are properly exposed, but high reflectivity objects become overexposed
Solution Approach 1:
The patent applies local quality by assigning different exposure time characteristics to different regions of the image based on object reflectivity. Low reflectivity objects (such as black objects) are assigned long exposure time data for proper exposure, while high reflectivity objects (such as white objects) are assigned short exposure time data to avoid overexposure. This localized approach ensures each object type receives optimal exposure treatment.
Solution Approach 2:
The patent changes the exposure time parameter dynamically based on the reflectivity characteristics of different objects in the scene. By calculating the proportion of overexposed and underexposed pixel points, the system adjusts the exposure time parameter to optimize the balance between capturing low reflectivity objects and preventing overexposure of high reflectivity objects.
3Object-affected harmful factors
If short exposure time is used to avoid overexposure, then high reflectivity objects are properly exposed, but low reflectivity objects become underexposed causing data loss
Solution Approach 1:
The patent merges long exposure time data and short exposure time data into a unified depth image by selectively combining pixel points from both datasets. The system merges the advantages of both exposure times: short exposure data for high reflectivity objects and long exposure data for low reflectivity objects, creating a composite image that avoids the limitations of using either exposure time alone.
Solution Approach 2:
The patent introduces an intermediary processing step that analyzes the proportion of overexposed and underexposed pixel points and uses this information to mediate between long and short exposure time data. This intermediary analysis enables the system to intelligently select and combine the most appropriate data from each exposure time dataset, resolving the contradiction between the two exposure approaches.
4Manufacturing precision
If exposure time is adjusted to optimize for one type of object, then that object type is properly exposed, but the other object type suffers from exposure issues
Solution Approach 1:
The patent achieves universality by creating a multi-functional exposure system that can handle both high reflectivity and low reflectivity objects within the same imaging process. The system simultaneously captures and processes both long and short exposure time data, enabling it to adaptively serve multiple object types with different reflectivity characteristics, rather than requiring separate optimization for each type.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method reduces pixel information loss and enhances image quality by optimizing exposure time based on pixel point proportions, resulting in improved depth image generation.
Implementation Method 1
Time of Flight (TOF) technology is one of the main technologies in 3D visual imaging technology
Data Source
AI summary
The present invention provides a method and system for processing an image. The method comprises: acquiring high-frame data and low-frame data of a target object, and acquiring initial exposure time to complete the high-frame data and the low-frame data; calculating a proportion of overexposed pixel points and a proportion of underexposed pixel points in the low-frame data; adjusting the initial exposure time according to a magnitude of the proportion of the overexposed pixel points and a magnitude of the proportion of the underexposed pixel points to obtain target exposure time; and acquiring target high-frame data and target low-frame data according to the target exposure time, and replacing overexposed pixel points in the target high-frame data with pixel points of the target low-frame data. The initial exposure time is adjusted according to the proportion of the overexposed pixel points to obtain the target exposure time, the target high-frame data and the target low-frame data are acquired according to the target exposure time, and the overexposed pixel points in the target high-frame data are replaced with the pixel points of the target low-frame data, thus the quality of a picture is improved.