Video Enhancement via Non-Linear Compression and Windowing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video enhancement methods for surveillance applications are computationally expensive and fail to provide sufficient resolution improvement, often resulting in lower image quality due to information loss from smoothing techniques, especially in images recreated from lossy compression, and introduce artifacts like 'ringing' in bright areas.
Innovation Solution
The implementation of a system that uses a non-linear compression algorithm and a two-dimensional windowing algorithm to reduce intensity values and prevent ringing artifacts, combined with frequency domain upsampling and edge detection, to enhance video resolution while maintaining image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If smoothing techniques are used to enhance video resolution, then image quality is improved, but information is lost especially in images recreated from lossy compression
Solution Approach 1:
The patent applies preliminary action by performing non-linear compression and windowing operations before the upsampling process. This preprocessing step prepares the image data in advance to prevent information loss during subsequent enhancement operations, particularly protecting bright areas from ringing artifacts while maintaining detail information for accurate reconstruction.
2Manufacturing precision
If conventional video enhancement methods are used, then resolution is improved, but computational expense increases and artifacts like ringing are introduced
Solution Approach 1:
The patent applies parameter changes by modifying the intensity values of pixels through non-linear compression and applying a two-dimensional windowing function. These parameter transformations change the distribution of pixel intensities to suppress ringing artifacts while maintaining edge sharpness, achieving high-resolution enhancement with reduced computational complexity compared to conventional methods.
3Object-generated harmful factors
If intensity values are compressed to prevent ringing artifacts, then artifact removal is improved, but image brightness may be affected
Solution Approach 1:
The patent applies local quality by using a two-dimensional windowing function that operates locally on the image data. This function selectively modifies intensity values in different regions of the image, particularly protecting bright areas from ringing artifacts while preserving the natural appearance of other regions. The local application of the windowing function ensures that brightness is maintained where needed while suppressing artifacts in problematic areas.
Data Source
AI summary
Embodiments of a method and apparatus for video enhancement are generally described herein. In some embodiments, the method includes accessing a first frame of the video. The first frame can include pixel data with intensity values within one of a number of non-overlapping ranges of intensity values. The method can further include compressing the intensity values of the first pixel data to generate non-linearly compressed (NLC) pixel data such that the intensity values are compressed into smaller ranges of intensity values. The method can further include upsampling the NLC pixel data based on an upsampling ratio. The method can further include decompressing the intensity values of the upsampled pixel data by reassigning intensity values of the upsampled pixel data to intensity values in the original intensity value ranges. Other example methods, systems, and apparatuses are described.