AI Object Recognition for Motion-Aware Noise Removal in Surveillance Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise removal technologies in surveillance cameras cause motion blur and increase image transmission bandwidth due to their inability to dynamically control noise removal intensity based on the presence or absence of objects and their movement in low-illumination environments.
Innovation Solution
A device and method for processing surveillance camera images that utilize artificial intelligence (AI) to dynamically control noise removal intensity. This involves recognizing objects using AI algorithms, calculating their moving speed, and adjusting noise removal intensity accordingly to minimize motion blur and noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If high sensor gain amplification is used in low-illumination environments, then image brightness is improved, but noise increases
Solution Approach 1:
The patent implements dynamic noise removal intensity control that adapts to object motion states. The system continuously monitors object presence and motion characteristics, then dynamically adjusts the noise removal intensity parameter in real-time. This dynamic adjustment allows the system to optimize the balance between noise removal and motion blur prevention based on current scene conditions.
Solution Approach 2:
The system changes the noise removal intensity parameter based on detected object motion characteristics. When stationary objects are detected, high noise removal intensity is applied. When moving objects are detected, the noise removal intensity is reduced to prevent motion blur. This parameter adaptation resolves the contradiction by making noise removal selective rather than uniform.
2Object-affected harmful factors
If high noise removal intensity is applied, then noise is reduced, but motion blur increases
Solution Approach 1:
The patent applies different noise removal intensities to different regions or objects based on their motion characteristics. Stationary background regions receive high noise removal intensity, while regions containing moving objects receive reduced or no noise removal processing. This local differentiation allows aggressive noise removal where safe and minimal processing where motion is present.
Solution Approach 2:
The system uses object detection and motion analysis results as feedback to control noise removal intensity. The object detection module continuously provides information about object presence and motion state, which feeds back to the noise removal module to adjust processing intensity. This closed-loop feedback mechanism ensures noise removal intensity is always appropriate for current scene dynamics.
3Shape
If noise removal intensity is decreased to reduce motion blur, then motion blur is reduced, but noise increases and bandwidth requirement increases
Solution Approach 1:
The system dynamically adjusts noise removal intensity based on real-time detection of object motion characteristics. Rather than using a fixed low intensity setting, the system continuously adapts the noise removal strength to match current scene conditions, applying higher intensity when appropriate and lower intensity when motion is detected.
Solution Approach 2:
The noise removal intensity parameter is changed based on object motion detection results. The system transitions between different intensity levels (high for stationary regions, low for moving regions) based on the detected motion state. This parameter adaptation allows optimal noise removal without excessive motion blur while managing bandwidth requirements.
4Measurement precision
If AI-based object recognition is implemented, then object recognition rate is improved, but device complexity increases
Solution Approach 1:
The patent divides the image processing task into separate functional modules: object detection module, motion analysis module, and noise removal control module. Each module performs a specific function and passes results to the next module. This segmentation allows the system to implement complex AI-based object recognition while maintaining manageable system architecture and processing flow.
Data Source
AI summary
Disclosed is a device for processing a camera image. The device for processing the camera image is configured to linearly control a noise removal intensity in accordance with a sensor gain amplification in an image obtained by an image capture device in a low luminance environment, and, when an object in the image is recognized, reduce the noise removal intensity in accordance with the speed of movement of the object, and thus, maintain a noise removal intensity having minimized motion blur.