Vibration-Compensated Blast Video Analysis for Automatic Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current blasting video analysis techniques are inefficient in handling dynamically changing shapes, colors, and textures, and are affected by camera vibration, making it difficult to automatically analyze and track high-speed particles and differentiate between smoke and dust clouds during blasting operations.
Innovation Solution
The proposed solution involves vibration-compensated background analysis to determine blast origin coordinates and estimate blast expansion trajectories, using vibration modeling to segment frame data into foreground and background binary images, and detecting toxic smoke based on video data, without the need for surface markers or high-speed cameras.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional computer vision algorithms are used for blast analysis, then object recognition works well for solid objects with well defined shapes, but the algorithms fail to handle blasts with dynamically changing shapes, colors and textures
Solution Approach 1:
The system transitions from static object recognition algorithms to dynamic background subtraction techniques that adapt to changing blast characteristics. The background model is continuously updated to accommodate dynamically changing shapes, colors and textures of blast clouds, allowing the system to track particles effectively despite their evolving properties.
Solution Approach 2:
The system changes the approach from recognizing fixed object parameters to tracking parameter changes over time. By monitoring temporal variations in pixel intensity, color, and position, the system identifies blast particles regardless of their changing characteristics, effectively adapting to dynamic blast conditions.
2Speed
If high-speed video recording is used to capture blast particles, then particle trajectories can be recorded, but strong camera vibration caused by blasts badly affects performance of conventional computer vision algorithms
Solution Approach 1:
The system extracts and removes the vibration component from the video analysis process. By separating the vibration-induced image variations from the actual particle motion signals, the system can reliably track particles even in the presence of strong camera vibration during high-speed recording.
Solution Approach 2:
The system introduces an intermediary vibration compensation mechanism that mediates between the vibrating camera and the analysis algorithms. This intermediary layer corrects for vibration effects before particle tracking, allowing reliable analysis despite camera instability during high-speed capture.
3Measurement precision
If manual tracking of particles with reticle is used, then trajectories can be plotted for analysis, but the process is not fully automatic and is time consuming
Solution Approach 1:
The system implements self-service automated particle tracking that eliminates manual intervention. The algorithm automatically detects particles, tracks their trajectories, and computes analysis results without requiring manual reticle operations, thereby maintaining measurement precision while dramatically improving analysis efficiency and productivity.
Solution Approach 2:
The system replaces the manual mechanical tracking process with an automated computational algorithm. Instead of using physical reticles and manual plotting, the system uses image processing algorithms to automatically detect and track particles, substituting mechanical manual operations with automated digital processing to enhance productivity.
4Object-affected harmful factors
If foam rubber mat is placed under camera tripod to reduce vibration, then some vibration effect is reduced, but it hardly eliminates strong vibration and makes camera setup less practical for industrial use
Solution Approach 1:
The system converts the harmful vibration effect into a beneficial signal by using the vibration characteristics themselves as part of the analysis. Instead of merely attempting to reduce vibration through physical dampening, the system processes the vibration-containing video data to extract useful particle trajectory information, making the setup more practical while utilizing the vibration data.
Data Source
AI summary
Blasting video analysis techniques and systems are presented using vibration compensated background analysis with automated determination of blast origin coordinates and highest point coordinates using blast outline coordinates for post-origin frames. Blast expansion trajectories of the highest particle are estimated in frames preceding the highest point frame, and estimated blast parameters including maximum height and initial blast velocity are computed independent of blast data.