Lens-Free Particle Counting via Trajectory Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Counting particles in a fluid, such as cells in a biological sample, is challenging due to the need to ensure each particle is counted only once despite appearing in multiple image frames during video capture.
Innovation Solution
A method and system using a lens-free imager to capture video data, where particle centers are linked across consecutive frames using a statistical model of particle displacements, and trajectories are tracked to calculate particle counts, either through linear regression or transformation into a common coordinate frame for clustering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video data is captured to track particles through multiple image frames, then particle identification and trajectory tracking are improved, but particle counting accuracy deteriorates due to multiple counting of the same particle
Solution Approach 1:
The system performs preliminary actions by capturing video data and detecting particle positions in multiple frames before counting. It establishes particle trajectories and determines effective observation volumes in advance, then uses this pre-established information to accurately count particles, preventing multiple counting of the same particle.
Solution Approach 2:
The system introduces an intermediary concept of 'effective observation volume' and particle trajectories as mediators between video capture and particle counting. By defining specific observation volumes and tracking particles through these volumes, the system accurately determines when a particle enters or exits the observation region, enabling precise counting without multiple counts.
2Device complexity
If conventional counting methods are used on video frames, then processing is simpler, but counting accuracy deteriorates due to inability to distinguish unique particles
Solution Approach 1:
The system transitions from two-dimensional frame-by-frame counting to three-dimensional trajectory-based counting by adding the time dimension. It tracks particle positions across multiple frames to establish trajectories, then counts particles based on their movement through defined observation volumes in space-time, fundamentally changing the counting approach from simple 2D detection to 3D trajectory analysis.
3Reliability
If particle tracking through multiple frames is implemented, then particle identification is improved, but processing time and computational complexity increase
Solution Approach 1:
The system segments the processing task into distinct phases: video capture and particle detection in each frame, trajectory construction by linking detections across frames, effective observation volume determination, and final counting based on trajectory analysis. This segmentation allows efficient processing by handling each phase separately with optimized algorithms for that specific task.
Data Source
Figure 1~2
Figure 3A~3B
Figure 4~5
AI summary
Systems and methods are provided for counting particles in a fluid flow. In an aspect, coordinates of particles are obtained from video data of particles in a fluid, the video data made up of a sequence of image frames. The particle positions are linked in each pair of consecutive image frames of the video data. The linked particle positions are used to calculate particle trajectories through sequential image frames of the video data, and the particles are counted based on the particle trajectory. In another aspect, the particle positons within each image frame are transformed to estimated positions within a common coordinate frame. The estimated particle positions of a particle are grouped into a cluster center, and the particle count is calculated based on the cluster centers.