Assembly Image Inspection Using Dynamic Frame Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image inspection technologies face challenges in determining defects in manual assembly processes in real-time, as they cannot guarantee correct component mounting and often require rework due to delayed inspections, and are not suited to variable assembly times.
Innovation Solution
An image inspection system with an analysis server and imaging devices that determine the start and end timings of assembly tasks, select appropriate inspection targets, and perform image inspections in real-time, allowing for immediate detection of defects during assembly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image inspection is performed after all components are mounted, then comprehensive defect detection is achieved, but large amount of rework is necessary
Solution Approach 1:
The system performs image inspection at intermediate stages during the assembly process, before all components are mounted. By conducting preliminary inspections at appropriate timings during assembly, defects are detected early when rework is minimal, rather than waiting until the end when comprehensive inspection would require extensive disassembly and rework.
Solution Approach 2:
The inspection timing is made dynamic and adaptable to the actual assembly process. The system determines appropriate inspection timings based on the assembly workflow, allowing inspections to be performed at optimal points during the process rather than at fixed intervals, thus balancing defect detection effectiveness with rework minimization.
2Loss of time
If inspection is performed early after each component mounting, then rework is reduced, but inspection timing cannot be fixed due to variable assembly times
Solution Approach 1:
The system incorporates feedback mechanisms that monitor the assembly process in real-time and dynamically adjust inspection timing based on actual workflow progress. By receiving feedback on assembly status and component mounting completion, the system determines optimal inspection moments adaptively, resolving the conflict between early inspection benefits and variable assembly timing constraints.
Solution Approach 2:
The system pre-defines multiple potential inspection timings corresponding to different assembly stages, and selects the appropriate timing based on real-time process monitoring. This preliminary preparation of multiple timing options allows the system to adapt to variable assembly durations while still performing inspections early enough to minimize rework.
3Reliability
If multiple inspection tasks are executed sequentially with substrate rotation and movement, then comprehensive inspection is achieved, but inspection timing cannot be synchronized with manual assembly variability
Solution Approach 1:
The system transitions from fixed sequential inspection tasks to a dynamic inspection approach that adapts to manual assembly variability. By making inspection timing and target selection dynamic based on real-time assembly status, the system maintains comprehensive inspection coverage while adapting to the unpredictable timing of manual component mounting operations.
Solution Approach 2:
The system pre-identifies multiple potential inspection targets and timings corresponding to different assembly stages, then selects the appropriate targets based on actual process progress. This preliminary preparation allows comprehensive inspection coverage to be achieved while adapting to the variable timing characteristics of manual assembly processes.
Data Source
AI summary
An image processing apparatus performs an inspection on an assembly component with use of one or more images of an assembling work, and includes a work determination unit configured to determine, from the one or more images, one or both of a start frame and an end frame of assembling work captured at a change point of the assembling work, a selection unit configured to select frames to be inspected from the one or more images based on a result of the determination by the work determination unit, and an inspection unit configured to perform an inspection on the frames selected by the selection unit.