Learned Feature Motion Detection for Low Power Camera Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion detection systems in camera hardware face challenges in accurately detecting motion with limited computing resources, often requiring downsampling of frames and reduced frame rates, which compromises processing efficiency and power consumption.
Innovation Solution
A data processing device and method that utilize a sampling unit, feature generation unit, and motion detection logic to form and compare current and learned features from sample points in image blocks, employing pseudo-random sampling and adaptive thresholding to determine motion, allowing for efficient motion detection at lower resolutions and power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If frames are downsampled and motion detection is performed at low frame rate, then processing burden and power consumption are reduced, but motion detection accuracy deteriorates
Solution Approach 1:
The image frame is divided into multiple blocks, and motion detection is performed independently on each block. This segmentation allows the system to process smaller regions with fewer computational resources while maintaining overall motion detection accuracy across the entire frame.
Solution Approach 2:
The patent applies different processing strategies to different blocks based on their characteristics. Blocks are classified into static and dynamic categories, allowing the system to allocate computational resources selectively - performing full motion detection on dynamic blocks while using simpler methods on static blocks, thereby improving accuracy without proportionally increasing power consumption.
2Measurement precision
If motion detection is performed on full-resolution frames at high frame rate, then motion detection accuracy is improved, but processing burden and device complexity increase
Solution Approach 1:
The patent divides the full-resolution frame into multiple smaller blocks for independent processing. This segmentation reduces the computational complexity per processing unit while maintaining the ability to detect motion accurately across the entire scene by aggregating results from all blocks.
Solution Approach 2:
The system performs motion detection selectively on certain blocks rather than uniformly processing all blocks with the same complexity. By identifying static blocks and applying simplified detection methods to them, the system reduces overall processing burden while maintaining accuracy in regions where motion is present.
3Use of energy by moving object
If limited computing resources are used for motion detection, then power consumption is reduced, but detection reliability deteriorates
Solution Approach 1:
The patent classifies blocks as static or dynamic and applies different detection strategies accordingly. This localized approach ensures reliable motion detection in dynamic regions while conserving energy in static regions, thereby maintaining overall detection reliability without proportionally increasing power consumption across the entire frame.
Data Source
AI summary
A data processing device for detecting motion in a sequence of frames each comprising one or more blocks of pixels, includes a sampling unit configured to determine image characteristics at a set of sample points of a block, a feature generation unit configured to form a current feature for the block, the current feature having a plurality of values derived from the sample points, and motion detection logic configured to generate a motion output for a block by comparing the current feature for the block to a learned feature representing historical feature values for the block.