Dynamic Frame Parameter Adjustment to Prevent Image Flicker
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital cameras experience the hunting phenomenon due to repeated adjustments in brightness or chrominance, leading to image flicker in modes like backlight and nightfall, where auto-exposure and auto-white balance continuously adjust settings without stabilizing.
Innovation Solution
A method to dynamically change frame parameters by determining if an input frame corresponds to a special scene mode or a general scene mode, predicting subsequent frames, and adjusting parameters accordingly to prevent repeated adjustments, thereby stabilizing settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If auto-exposure continuously adjusts brightness to adapt to changing lighting conditions, then adaptability is improved, but image flicker occurs due to repeated adjustments
Solution Approach 1:
The patent applies preliminary anti-action by predicting the scene mode of the next frame in advance and pre-adjusting the parameter accordingly. When a special scene mode is detected, the parameter is adjusted before the hunting phenomenon can occur, preventing the back-and-forth oscillation. The prediction mechanism anticipates lighting changes and prepares the parameter adjustment beforehand, counteracting the potential flicker effect.
Solution Approach 2:
The patent implements preliminary action by determining the scene mode and adjusting parameters based on predicted future conditions rather than reacting to past changes. The system analyzes the current frame, predicts the next frame's scene mode, and proactively sets the parameter to an appropriate value before the actual lighting change occurs, thereby avoiding repeated adjustments and image flicker.
2Measurement precision
If auto-white balance continuously modifies chrominance to adapt to different lighting temperatures, then color accuracy is improved, but hunting phenomenon occurs due to repeated chrominance adjustments
Solution Approach 1:
The patent applies preliminary anti-action by predicting the scene mode of the next frame and pre-adjusting the chrominance parameter accordingly. When a special scene mode requiring white balance adjustment is detected, the system prepares the chrominance value in advance, preventing the repeated back-and-forth adjustments that cause hunting phenomenon while maintaining color accuracy.
Solution Approach 2:
The patent implements preliminary action by determining the scene mode and setting the chrominance parameter based on predicted future conditions. The system analyzes the current frame, predicts the next frame's scene mode, and proactively adjusts the white balance parameter before the actual lighting temperature change occurs, thereby avoiding repeated chrominance adjustments and stabilizing the image.
3Speed
If the parameter is adjusted frequently to respond to scene changes, then responsiveness is improved, but hunting phenomenon is generated due to oscillating adjustments
Solution Approach 1:
The patent implements preliminary action by predicting the scene mode of the next frame and adjusting the parameter proactively before the scene change actually occurs. This approach maintains high responsiveness to scene changes while avoiding the oscillating adjustments that cause hunting phenomenon, as the parameter is set to the appropriate value in advance rather than reacting repeatedly to each change.
Solution Approach 2:
The patent employs feedback by using the predicted scene mode information to control parameter adjustments. The system continuously monitors scene mode changes and uses this feedback to make intelligent decisions about when and how to adjust parameters, preventing unnecessary oscillations while maintaining appropriate responsiveness to actual scene changes.
Data Source
AI summary
A method of dynamically changing a parameter of a frame is disclosed. An input frame is determined as a special scene or a general scene. A subsequent frame is predicted by changing the parameter of the subsequent frame according to a special-scene operation when the input frame is the special scene, and by changing the parameter of the subsequent frame according to a general-scene operation when the input frame is not the special scene. The predicted frame is determined whether to be a special scene. Accordingly, the parameter of the input frame is changed according to the special-scene operation when both the input frame and the predicted frame are the special scenes; the parameter of the input frame is changed according to the general-scene operation when both the input frame and the predicted frame are not the special scenes; and the parameter of the input frame is maintained without change when only one of the input frame and the predicted frame is the special scene.