Object Detection Apparatus Using Segmented Processing Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar techniques face challenges in simultaneously improving detection accuracy for various objects, such as vehicles, pedestrians, and roadside objects, due to differing detection requirements and increased processing loads from noise, leading to mis-extraction of peaks.
Innovation Solution
The object detection apparatus performs individualized signal processing for each type of object, using a signal processing unit to analyze beat signals and an object detection unit that executes extracting, tracking, and recognition processes tailored to the object's features, preventing unnecessary candidate extraction and optimizing processing based on object characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a unified detection process is used for all objects, then the system complexity is low, but detection accuracy for different object types cannot be simultaneously improved
Solution Approach 1:
The detection process is segmented into multiple independent processing lines, each dedicated to a specific object type (vehicle, pedestrian, cyclist). Each processing line applies object-specific extraction criteria, tracking parameters, and recognition rules, enabling optimized detection accuracy for each category without requiring a single complex unified process
Solution Approach 2:
The system dynamically selects and applies different processing configurations based on the detected object type. Each object category has its own customizable processing parameters that can be independently adjusted, allowing the system to adapt its detection strategy to the specific characteristics of each object type
2Reliability
If all peaks are extracted without missing any, then detection coverage is complete, but processing load increases and peaks are likely to be mis-extracted due to noise
Solution Approach 1:
Different extraction criteria and threshold settings are applied locally to each object type's processing line. For example, vehicle detection may use different peak extraction thresholds than pedestrian detection, allowing each object category to be detected with optimized parameters that reduce noise interference while maintaining detection coverage
Solution Approach 2:
The system changes processing parameters (such as peak extraction thresholds, tracking windows, and recognition criteria) based on the object type being detected. This allows the system to maintain high detection coverage for each object category while reducing false positives and processing load through object-specific parameter optimization
Data Source
AI summary
The object detection apparatus is provided with a signal processing unit and an object detection unit. The signal processing unit performs a frequency analysis of a beat signal obtained by transmitting and receiving continuous waves and estimates an incoming direction of reception waves, the object detection unit executes, based on a processing result of the signal processing unit, at least an extracting process of an object candidate, a tracking process for an object and the object candidate, and an object recognition process that recognizes the object candidate to be the object.The object detection unit is characterized in that the unit executes, individually for each type of object to be detected, the extracting process, the tracking process and the object recognition process, and changes content of processes depending on a feature of the object to be detected.