Anomalous Pixel Correction in Time of Flight Depth Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for correcting anomalous pixels in depth maps from Time of Flight cameras are inefficient and reduce the number of pixel points, leading to incomplete images and ineffective edge handling.
Innovation Solution
A method that calculates a matching index of pixels between two point cloud maps, identifies anomalous pixels based on a correction threshold, and corrects them using a correction column difference, thereby retaining pixel integrity and improving correction efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If traditional filtering methods are used to remove flying-spot noise, then noise reduction is achieved, but pixel points are reduced leading to image incompleteness
Solution Approach 1:
The patent changes the approach from binary filtering (remove/keep) to continuous parameter adjustment by modifying pixel depth values. Instead of removing anomalous pixels, the method adjusts their depth parameters based on matching indices and correction column differences, transforming harmful noise into valid data while preserving pixel quantity.
Solution Approach 2:
The patent recovers discarded information by identifying that traditional methods discard anomalous pixels, but these pixels contain recoverable valid information. The method recovers this information by calculating matching indices between stereo images and using correction column differences to restore the original depth values, thus recovering pixel points that would otherwise be lost.
2Measurement precision
If complex judgment and calculation methods are used for pixel correction, then correction accuracy may be improved, but correction efficiency decreases
Solution Approach 1:
The patent extracts only the essential information needed for correction: matching indices between stereo images and correction column differences. By taking out only these critical parameters rather than performing complex multi-step analysis, the method achieves both accuracy and efficiency, avoiding unnecessary computational overhead.
Solution Approach 2:
The correction process is segmented into distinct independent steps: first calculating matching indices, then determining correction column differences, and finally applying corrections. This segmentation allows each step to be optimized independently and processed efficiently, improving overall productivity while maintaining precision.
Data Source
AI summary
A method for correcting the anomalous pixel includes: calculating a matching index of pixels in a first point cloud map and a second point cloud map; finding out an anomalous pixel based on the matching index of the pixels and a correction threshold; and calculating a correction column difference, and correcting the anomalous pixel based on the correction column difference and the matching index of the pixels. The anomalous pixel in the point cloud map is retained and corrected, thus the pixel integrity of an image is guaranteed; and a mode for correcting the anomalous pixel is simple, requires fewer calculating steps, and exhibits high anomalous pixel correction efficiency. An apparatus for the same purpose includes a first camera, a second camera, an image processing unit, an indexing unit, a calculating unit, a judging unit, and a correction execution unit.