Vehicle Image Processing Rate Control for Object Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems struggle to perform appropriate processing on subjects like pedestrians or fast-moving objects due to increased frame rates affecting processing time and positional changes, leading to inadequate recognition.
Innovation Solution
An image processing device that adjusts the number of images processed per unit time based on the category and state of the object, using a recognition unit to categorize objects and a decision unit to set the processing rate accordingly, allowing for appropriate image processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the frame rate is increased to improve detection capability when the vehicle may be at risk, then the detection sensitivity is improved, but the image processing time per frame is shortened and high calculation cost processing cannot be performed
Solution Approach 1:
The image processing device dynamically adjusts the frame rate based on the detected object category. When a pedestrian is detected, the system switches to a lower frame rate (e.g., 30 fps or lower) to allocate sufficient processing time for high calculation cost operations. When no pedestrian or other objects are detected, the system uses a higher frame rate (e.g., 60 fps) to maintain detection sensitivity. This dynamic adjustment resolves the contradiction by adapting the frame rate to the specific processing requirements of different object categories.
Solution Approach 2:
The system changes the frame rate parameter based on the object category. By identifying the subject type (pedestrian, vehicle, animal, etc.), the device modifies the processing parameters accordingly. For pedestrians requiring high calculation cost processing, the frame rate is reduced to ensure adequate processing time. For other objects, the frame rate is maintained at higher levels. This parameter change strategy allows the system to optimize between detection sensitivity and processing time allocation.
2Reliability
If the frame rate is increased to improve detection capability, then the number of frames input per unit time increases, but the position of the subject greatly changes from one frame to the next affecting difference detection
Solution Approach 1:
The system dynamically adjusts the frame rate based on the detected object category to optimize difference detection. For moving objects like vehicles where position changes significantly, the system uses a higher frame rate to capture more intermediate positions and improve motion tracking accuracy. For stationary or slow-moving objects like pedestrians, a lower frame rate is sufficient and reduces positional variation between frames. This dynamic adaptation resolves the contradiction by matching the frame rate to the motion characteristics of different object categories.
3Stability of the object's composition
If a standard frame rate is maintained to ensure consistent processing, then the processing stability is improved, but the processing cannot be optimized for different subject types
Solution Approach 1:
The image processing device implements dynamic frame rate adjustment based on object category detection. The system maintains processing stability through a standardized detection framework while adapting the frame rate parameter according to the specific subject type. When pedestrians are detected, the system switches to optimized lower frame rates for high calculation cost processing. When other objects are detected, different frame rates are applied. This dynamic adaptation enables both stability through consistent processing architecture and versatility through category-specific optimization.
Solution Approach 2:
The system changes the frame rate parameter based on the detected object category while maintaining a stable processing pipeline. The detection unit identifies object categories (pedestrian, vehicle, animal, etc.) and the control unit adjusts the frame rate accordingly. This parameter change allows the system to optimize processing for different subject types while maintaining overall processing stability through a unified control mechanism.
Data Source
Figure 1
Figure 2A~2D
Figure 3
AI summary
An image processing device (1) includes a recognition unit (10) configured to recognize a category of an object (8, 9) within a predetermined distance with respect to a vehicle (2) based on a detection result of a sensor (3), the sensor (3) including an image sensor, a decision unit (11) configured to decide the number of images subjected to image processing per unit time based on the category of the object (8, 9) recognized by the recognition unit (10), an acquisition unit (12) configured to acquire, from the image sensor, a plurality of continuously shot images with the object as a subject, and a calculation unit (13) configured to select a target image from the images acquired by the acquisition unit (12) based on the number of images subjected to the image processing per unit time decided by the decision unit (11), and to calculate information regarding the object (8, 9), based on the target image.