Vision Detector High-Frame-Rate Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine vision systems in industrial manufacturing face limitations due to their reliance on single-image analysis, slow operation speeds, and the need for trigger signals, leading to incorrect decisions and complex setup processes, especially when dealing with objects in continuous motion.
Innovation Solution
The Vision Detector Method and Apparatus employ high-frame-rate image capture and analysis, using multiple perspectives and dynamic image analysis to detect objects without triggers, with a vision detector that captures and analyzes sequences of images at rates up to 500 frames per second, enabling reliable detection and inspection of objects in motion with synchronized output signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single image is captured and analyzed for each object, then the vision system operates simpler and faster, but the decision reliability deteriorates due to limited viewing perspectives
Solution Approach 1:
The patent segments the inspection process into multiple independent image captures from different positions, where each image provides a separate viewing perspective. The system divides the continuous motion trajectory into discrete capture points, allowing each segment to be analyzed independently while collectively providing comprehensive inspection coverage that improves decision reliability without requiring a single complex multi-position system.
Solution Approach 2:
The patent transitions from a single two-dimensional image to a three-dimensional inspection by capturing images at multiple positions along the motion trajectory. This adds the temporal/dimensional aspect of movement, creating a volumetric understanding of the object that enables more reliable detection of defects and features that would be invisible from a single perspective.
2Reliability
If a trigger signal is used to detect objects in continuous motion, then the detection reliability improves, but the device complexity and setup difficulty increase
Solution Approach 1:
The patent implements self-triggering where the vision system automatically detects objects by monitoring changes in the captured images themselves, eliminating the need for external trigger signals. The system uses image analysis to detect when an object enters the field of view and automatically initiates capture and analysis sequences, making the system self-sufficient and reducing setup complexity while maintaining detection reliability.
Solution Approach 2:
The patent employs feedback mechanisms where the system continuously monitors image data for object presence and uses this information to control the capture timing and analysis processes. This closed-loop feedback eliminates the need for separate trigger signals by using the visual information itself to regulate system operation, reducing complexity while maintaining reliable object detection.
3Reliability
If the vision system operates at high frame rates to capture multiple perspectives, then the detection reliability improves, but the processing time and computational load increase
Solution Approach 1:
The patent segments the high-frame-rate video stream into discrete, meaningful frames based on object presence and motion detection. Rather than processing every frame uniformly, the system identifies key frames that contain critical inspection information and focuses computational resources on analyzing these selective frames, reducing overall processing time while maintaining the benefits of high-frame-rate capture for reliable detection.
Solution Approach 2:
The patent applies partial analysis to a subset of captured frames rather than processing all frames in detail. By using motion detection and object presence identification to select only the most relevant frames for comprehensive analysis, the system achieves reliable multi-perspective inspection while reducing computational load and processing time compared to analyzing every captured frame.
4Manufacturing precision
If traditional machine vision systems are used with single-image analysis, then the setup is simpler, but incorrect decisions occur more frequently due to limited viewing perspectives
Solution Approach 1:
The patent implements self-triggering where the vision system automatically detects objects by monitoring changes in the captured images themselves, eliminating the need for external trigger signals. The system uses image analysis to detect when an object enters the field of view and automatically initiates capture and analysis sequences, making the system self-sufficient and reducing setup complexity while maintaining detection reliability.
Solution Approach 2:
The patent segments the inspection process into multiple independent image captures from different positions, where each image provides a separate viewing perspective. The system divides the continuous motion trajectory into discrete capture points, allowing each segment to be analyzed independently while collectively providing comprehensive inspection coverage that improves decision reliability without requiring a single complex multi-position system.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for faster and more reliable detection and inspection of objects in motion, reducing incorrect decisions and setup complexity, while being less expensive and easier to configure than traditional machine vision systems.
Implementation Method 1
Each pixel measures the intensity of light falling on it during the shutter time
Implementation Method 2
The image is captured by exposing a two-dimensional array of photosensitive elements for a brief period, called the integration or shutter time, to light that has been focused on the array by a lens
Data Source
AI summary
Disclosed are systems and methods for setting various operating parameters of a vision detector from production line information that can be supplied by a manufacturing technician who is not skilled in the art of the invention. These operating parameters include shutter time, video gain, idle time, frame count, and locator search range. The production line information includes line speed, field of view size, direction of motion, and object spacing.