OCR Wire Harness Matching with Geometric Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for assembling complex wire harnesses into connectors are prone to errors due to small wire diameters, numerous wires, close proximity, and limited workspace, making precise pinning operations challenging and inefficient.
Innovation Solution
An optical character recognition (OCR) system that uses geometric modeling for character recognition, restricts searches by region and character, and employs a compact computer-based system with an LED light panel to guide wire placement, allowing for precise registration and accurate assembly in confined areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional manual wire pinning methods are used, then the assembly process is simple and equipment-free, but the error rate increases due to small wire diameters, numerous wires, close proximity, and limited workspace
Solution Approach 1:
The patent replaces manual mechanical wire pinning operations with an automated optical system. A camera captures images of wire labels, OCR software automatically reads and identifies wire numbers, and the system electronically displays corresponding pin locations, substituting human visual inspection and manual matching with automated optical and computational systems.
Solution Approach 2:
The patent introduces an intermediary computational system that acts as a mediator between the physical wire harness and the connector. The system includes a computer with OCR software that processes images of wire labels, matches them against a database of wire-to-pin mappings, and displays the results on a monitor, serving as an intelligent intermediary that eliminates direct human intervention in the matching process.
2Productivity
If the technician manually locates pin positions in the connector, then no additional equipment is needed, but the operation becomes time-consuming and error-prone in confined spaces
Solution Approach 1:
The patent replaces the manual mechanical process of physically searching for and locating pin positions in the connector with an automated visual display system. The computer system electronically presents pin location information on a monitor, allowing the technician to view pin positions remotely without physically navigating the confined connector space, thereby improving both speed and ease of operation.
3Measurement precision
If wire labels are read manually by the technician, then the process requires minimal equipment, but reading accuracy decreases due to small label size and poor lighting conditions
Solution Approach 1:
The patent replaces manual visual reading of wire labels with an automated optical character recognition system. A camera captures images of the wire labels, and OCR software automatically identifies and reads the wire numbers, eliminating the need for the technician to visually inspect small labels in potentially poor lighting conditions. This substitution significantly improves reading accuracy while the system's lighting requirements are minimal, as the camera can capture labels with standard illumination.
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
The system enables fast, accurate, and error-free assembly of wire harnesses with connectors, even in remote and limited-access environments, by using OCR to identify wire numbers and illuminate the correct pin locations, improving efficiency and reducing human error.
Implementation Method 1
An optical character recognition (OCR) system that uses geometric modeling for character recognition, restricts searches by region and character, and employs a compact computer-based system with an LED light panel to guide wire placement
Data Source
AI summary
An OCR system for matching wire harnesses and connectors facilitates precise registration of wire number strings, uses geometric modeling for character recognition, and restricts searches by region and character to ensure speed and accuracy. A string location algorithm is used to search for and identify the location of the beginning of a wire number string. The horizontal edges of the wire in the image are located, a diameter of the wire is determined, light intensity is confirmed, and the first character is found. The resulting coordinate is used by the algorithm for character definition. Geometric shapes are used for identification in order to overcome twisted wires, poorly printed markings, ink color variations, and contacting characters.