Image Transmission Map Optimizer for Haze Compensation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional image processing methods for compensating image transmission rate in outdoor environments, such as those affected by smog or haze, often result in erroneous transmission map estimations due to object color and brightness, leading to unsatisfactory de-hazing effects.

Innovation Solution

An image processing apparatus comprising an image transmission map estimator, a transmission map optimizer, and an image rebuilder, which performs smoothing operations on estimated transmission maps to generate optimized maps, and includes a de-hazing strength generator to dynamically adjust de-hazing strengths based on image hazy status, enhancing image quality by rebuilding the input image.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional transmission map estimation is performed on input images, then de-hazing compensation is attempted, but the estimation becomes erroneous due to object color and brightness characteristics

Engineering Contradiction:
Improvetransmission map estimation accuracyVSAvoidde-hazing effect reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces a color space conversion intermediary mechanism that transforms the transmission map from the original color space to a transformed color space. This intermediary conversion process eliminates the erroneous estimations caused by object color and brightness characteristics, thereby improving both measurement precision and reliability of the de-hazing effect.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Manufacturing precision

If transmission map optimization with multiple smoothing operations is applied, then image clarity is improved, but processing time and computational complexity increase

Engineering Contradiction:
Improveimage clarityVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent implements a dynamic optimization mechanism where the transmission map optimizer selectively applies smoothing operations based on image characteristics and hazing severity. The system dynamically adjusts the number and strength of smoothing operations required, rather than applying a fixed number of operations, thereby reducing processing time while maintaining image clarity.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameters of the transmission map through multiple smoothing operations with different strengths. By adjusting these parameters adaptively, the system achieves optimal image clarity without unnecessarily increasing processing time, as the parameter changes are driven by actual image analysis rather than fixed protocols.

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If de-hazing strength is dynamically adjusted based on image hazy status, then image quality enhancement is optimized, but system complexity increases

Engineering Contradiction:
Improveimage qualityVSAvoidsystem complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent implements a feedback mechanism where the de-hazing strength generator continuously monitors image hazy status and dynamically adjusts de-hazing strength accordingly. This feedback loop allows the system to optimize image quality enhancement while managing complexity through automated adaptive control rather than manual intervention.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS10043246B2Image processing apparatus
Publication Date: 2018.08.07 NOVATEK MICROELECTRONICS CORP

AI summary

An image processing apparatus including an image transmission map estimator, a transmission map optimizer, and an image rebuilder is provided. The image transmission map estimator receives an input image and estimates transmission rate of the input image to generate an estimated transmission map. The transmission map optimizer receives the estimated transmission map and operates smooth operations with different strength on the estimated transmission map to respectively generate a plurality of smoothed transmission maps. The transmission map optimizer generates an optimized transmission map according to the estimated and smoothed transmission maps. The image rebuilder receives the optimized transmission map and generates an output image by rebuilding the input image according to the optimized transmission map.