Pixel Density Compensation for Wireless Video Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video analytics systems for Public Safety applications lack a mechanism to maintain a minimum pixel density across an object of interest within video transmitted over a wireless network, considering factors like field of vision and network throughput.
Innovation Solution
A method and apparatus that utilize a video analytic device to determine the current pixel density of an object of interest, compare it to a minimum required pixel density, and adjust either the camera's zoom feature or encoded frame resolution to ensure the object maintains the necessary pixel density for effective video transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the encoded frame resolution is increased to maintain minimum pixel density, then the pixel density of the object of interest is improved, but the network throughput requirement increases
Solution Approach 1:
The system dynamically adjusts the encoded frame resolution based on real-time network conditions. When network throughput is sufficient, higher resolution is used to maintain minimum pixel density for accurate object identification. When network throughput is limited, the system reduces resolution while maintaining adequate pixel density through adaptive bitrate streaming, thus resolving the contradiction between image quality and network bandwidth consumption.
Solution Approach 2:
The system changes the encoding parameters (resolution, bitrate, frame rate) based on network conditions and object characteristics. By analyzing the object's size and position in the frame, the system adjusts encoding parameters to ensure minimum pixel density is maintained for critical objects while optimizing overall throughput, effectively managing the trade-off between quality and bandwidth.
2Area of stationary object
If the field of vision is increased to maximize situational awareness, then the coverage area is improved, but the pixel density of the object of interest decreases
Solution Approach 1:
The system divides the video stream into multiple segments or layers (e.g., base layer and enhancement layers in hierarchical video coding). The base layer provides low-resolution overview for situational awareness, while enhancement layers provide high-resolution details for object identification. This segmentation allows simultaneous optimization of both field of vision coverage and pixel density for objects of interest.
Solution Approach 2:
The system applies different quality levels to different regions of the video frame. Areas with objects of interest receive higher encoding quality and resolution to maintain minimum pixel density, while areas without objects of interest use lower quality to maximize overall field of vision coverage. This local quality adjustment resolves the contradiction between comprehensive coverage and detailed observation.
Data Source
AI summary
A method includes receiving a frame containing an object of interest and determining a current pixel density of the object of interest. The current pixel density is compared to a minimum pixel density to generate a comparison result and an indication is received of available throughput of a wireless network used to transmit the video. Based on the comparison result and the available throughput, a pixel density compensation method is selected from multiple methods, which include adjusting a zoom feature of a camera used to capture the video and adjusting encoded frame resolution. Using the selected pixel density compensation method, a pixel density compensation parameter is determined to adjust the current pixel density with respect to the minimum pixel density for the object of interest. The pixel density compensation parameter is provided to one or both of the camera or an encoder that sends the video over the wireless network.