Hierarchical Pyramid Object Detection Using Spatio-Temporal Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current object detection systems in applications like automatic driving and unmanned aerial vehicles face challenges in accurately detecting and tracking objects over time, as they primarily rely on spatial information without effectively incorporating time information, which limits their detection performance.
Innovation Solution
The system generates pyramid images from input images captured at different times, adding time information to these images, and uses deep learning based on neural networks to extract and track objects, improving detection performance by utilizing both spatial and temporal data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only spatial information from single images is used for object detection, then the detection process is simple and fast, but the detection accuracy and tracking performance deteriorate due to lack of temporal context
Solution Approach 1:
The patent segments the detection process by generating multiple pyramid images from different time points and processing them separately through the object extractor, which then combines results. This segmentation of temporal data into discrete processable units improves accuracy without overwhelming system complexity
Solution Approach 2:
The patent adds the time dimension to traditional spatial object detection by capturing images at multiple time points and generating pyramid images across temporal sequences. This transforms the detection from 2D spatial analysis to 3D spatio-temporal analysis, improving detection accuracy through temporal context
2Reliability
If multiple images captured at different times are processed to add time information, then object tracking performance improves, but processing time and computational load increase
Solution Approach 1:
The patent performs preliminary actions by pre-processing images to generate pyramid structures before object extraction. By organizing images into pyramid formats at different time points in advance, the system prepares temporal data for efficient processing, improving tracking reliability while managing processing time through structured pre-computation
Solution Approach 2:
The patent creates pyramid image copies from original captured images at different time points. These copied and transformed pyramid images serve as processed representations that preserve temporal information while enabling efficient object extraction and tracking without requiring continuous access to raw high-resolution images
Data Source
AI summary
An object detecting system, which detects an object by using a hierarchical pyramid structure includes: a pyramid image generator configured to receive a plurality of input images respectively corresponding to a plurality of time points and generate a plurality of pyramid images corresponding to each of the plurality of input images; an object extractor configured to generate a plurality of pieces of object data by extracting at least one object from the plurality of pyramid images; and a buffer storing the plurality of pieces of object data on an object basis.