Object Detection Region Adjustment via Orientation Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object detection systems face challenges in reducing processing load while maintaining accurate detection, particularly when the orientation of a moving object changes, leading to inefficiencies and potential misses in detecting target objects.
Innovation Solution
An information processing device that adjusts the processing target area for object detection based on sensor data from ambient environment and orientation sensors, dynamically determining the area to focus on for image recognition processing, thereby optimizing the processing load and reducing the risk of missing detection targets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the detection target region is preset based on top contour images of vehicles, then the amount of processing for object detection is reduced, but the detection accuracy deteriorates when the orientation of the moving object changes
Solution Approach 1:
The patent applies the dynamics principle by making the detection target region adjustable based on the orientation of the moving object. Instead of using a fixed preset region, the system dynamically modifies the detection region according to sensor data about the device's orientation, allowing the processing area to adapt to changing conditions while maintaining reduced computational load.
Solution Approach 2:
The patent implements parameter changes by modifying the detection target region parameters (position, size, shape) based on orientation sensor data. The system changes the geometric parameters of the processing area to match the actual orientation of the moving object, thereby maintaining detection accuracy without processing the entire image area.
2Loss of time
If the processing target area is reduced to decrease computational load, then the processing time is reduced, but the risk of missing detection targets increases
Solution Approach 1:
The patent applies preliminary action by obtaining orientation sensor data before performing object detection. This allows the system to pre-calculate and set the appropriate detection target region based on the current orientation, ensuring that the reduced processing area is optimally positioned to capture potential detection targets before the actual detection process begins.
Solution Approach 2:
The system uses feedback from orientation sensors to continuously adjust the detection target region. The sensor data provides real-time information about the device's orientation, which is fed back to modify the processing area parameters, creating a closed-loop system that adapts to changing conditions while maintaining detection reliability.
3Measurement precision
If contour detection is performed as preprocessing for object detection, then object detection accuracy is improved, but the amount of processing increases significantly
Solution Approach 1:
The patent extracts only the essential preprocessing step needed for accurate detection within the reduced target region. Instead of performing full contour detection on the entire image, the system extracts and processes only the relevant portion of the image defined by the orientation-based detection region, thereby maintaining necessary detection precision while significantly reducing processing complexity.
Data Source
AI summary
An information processing device includes a processor. This processor: obtains at least one of first sensor data output from a first sensor and used to determine an ambient environment of a device in which a third sensor is placed, and second sensor data used to determine an orientation of this device; determines a tilt of a plane in a sensing direction of the third sensor with respect to the orientation of the device based on the at least one of the first sensor data and the second sensor data; determines, in accordance with the tilt determined, a processing target area of third sensor data output from the third sensor and used for object detection processing in the sensing direction of the third sensor; and executes the object detection processing using the processing target area determined, of the third sensor data.