Block-Based Histogram of Oriented Gradients for Real-Time Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional object detection systems require significant computing resources and have slow detection speeds, making them unsuitable for real-time applications like vehicle navigation, where instantaneous object detection is crucial, especially in scenarios like traffic sign recognition.
Innovation Solution
A processor-implemented method using a block-based histogram of oriented gradients approach, where images are divided into blocks, and magnitude and angle values are calculated and normalized, allowing for faster processing by comparing these values with specific feature vectors to detect objects like traffic signs, vehicles, and pedestrians using fewer computing resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional object detection systems are implemented using software or FGPA, then object detection functionality is achieved, but computing resource consumption increases and detection speed decreases
Solution Approach 1:
The image is divided into multiple blocks, and each block is processed independently to calculate gradient magnitudes and orientations. This segmentation enables parallel processing of different regions, significantly improving detection speed while reducing the computational burden on each processing unit. The blocks are further divided into cells, allowing hierarchical processing that optimizes resource utilization.
Solution Approach 2:
The patent transitions from software-based or FGPA-based detection to a hardware implementation using an array of processing elements organized in a dimensional structure. This hardware dimensionality change enables simultaneous execution of multiple gradient calculations across different image blocks, achieving real-time detection performance with reduced computing resource consumption per detection operation.
2Speed
If conventional object detection systems are implemented using software, then flexibility is maintained, but detection speed is slow and cannot meet real-time requirements
Solution Approach 1:
The patent replaces the software-based detection mechanism with a hardware-based system consisting of an array of processing elements. This substitution of the detection mechanism from software to hardware enables parallel execution of gradient calculations and histogram computations, achieving real-time detection speed while managing system complexity through structured hardware organization.
3Productivity
If FGPA is used for object detection, then real-time performance is achieved, but the system becomes inflexible and difficult to integrate with vehicle electronics
Solution Approach 1:
The processing elements in the array are designed with universal functionality to handle multiple object types including traffic signs, vehicles, and pedestrians. Each processing element can compute gradient histograms for different block configurations, making the system adaptable to various detection scenarios. This multi-functionality is achieved through configurable processing logic that can be programmed for different detection requirements, providing integration flexibility with vehicle electronics.
Data Source
AI summary
A processor-implemented method and system for detecting objects in a media file using a block based histogram of oriented gradients methodology is described. At least one image representing at least one frame of a video sequence comprising one or more objects of at least one desired object type is received. A region of interest contained in a block of frame is considered. An array containing magnitude and angle values is calculated, normalized for each pixel in the block in the internal memory. The normalized array is aggregated, ported and stored in an external memory. A value or block of values of the aggregated array in the external memory along with the specific feature vector is transferred to the internal memory. Based on a comparison of each value in the aggregated array and the specific feature vector, a desired object type is detected.