Dual Object Detector System for Real-Time Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object detection systems struggle to accurately determine if a detected object belongs to a specific class, especially when the object is partially occluded in the image frame, leading to increased processing time and reduced efficiency in real-time applications.
Innovation Solution
The method involves using a first object detector with higher precision but longer processing time, and a second object detector with lower precision but shorter processing time, performing object detection in parallel. The second object detector's results are used to enhance the first object detector's results, allowing for more accurate classification even when the object is partially occluded.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an enhanced object detector is used to determine object class even when partially occluded, then detection accuracy is improved, but processing time increases
Solution Approach 1:
The object detection system is segmented into two separate detectors: a first object detector that operates with higher precision for accurate classification and a second object detector that operates with lower precision but faster processing speed. This segmentation allows each detector to be optimized for its specific function, resolving the contradiction between accuracy and speed.
Solution Approach 2:
A second object detector is introduced as a simpler copy or variant of the first object detector. This second detector replicates the basic object detection functionality but with reduced complexity to achieve faster processing times, while the first detector provides the enhanced accuracy when needed.
2Measurement precision
If a single high-precision object detector is used, then detection accuracy is improved, but processing time increases and frame rate decreases
Solution Approach 1:
The detection system is divided into two functional components: the first object detector handles accuracy-critical tasks with higher precision, while the second object detector handles speed-critical tasks with lower precision. This segmentation enables the system to maintain high frame processing rates by using the faster second detector for routine detections and reserving the slower first detector for cases requiring high accuracy.
Solution Approach 2:
The system changes the operational parameters of the two detectors differently: the first detector uses higher precision parameters (slower processing) while the second detector uses lower precision parameters (faster processing). This parameter differentiation allows the system to optimize productivity by selecting the appropriate detector based on the specific detection requirements.
3Reliability
If object detection is performed on each image frame, then detection completeness is improved, but processing time increases
Solution Approach 1:
The system applies partial action by using the second object detector for routine frames where high precision is not critical, and reserves the first object detector for specific frames where detection completeness is paramount. This partial application of high-precision detection only where necessary maintains reliability while reducing overall processing time.
Data Source
AI summary
A method for classifying a detected object is disclosed. First and second object detectors detect first and second objects in first and second image frames, respectively, of a video sequence, and first and second probability scores respectively are calculated indicating a probability that the detected object belongs to a specific class. The second image frame is subsequent to the first image frame. The first object detector has a higher object detection precision and a longer processing time than the second object detector. The first and second object detections are performed in parallel. Reducing the first classification threshold or increasing the first probability score are performed if the first probability score is below a first classification threshold and the second probability score is above a second classification threshold. The first object is determined to belong to the specific class based on the probability scores and the classification thresholds.