Deep Machine Vision Platform (ART/240CP)

Deep Machine Vision Platform (ART/240CP)

  • Deep Machine Vision Platform (ART/240CP)
    ART/240CP
    Platform
    16 / 11 / 2017 - 15 / 11 / 2019
    14,721

    Dr Chen-jung TSAI

    1. Deep machine vision software platform: Vision measurement software a. GUI based IDE for vision application R&D b. Runtime processing module for vision application deployment c. 2D/3D vision measurement module d. Computer aided parameter tuning module Specifications: Runtime environment: Windows / Linux / ARM Development environment: Windows Speed: < 0.2s for template matching @4M camera I/O: Serial port / TCP / UDP / GPIO 2. Deep machine vision software platform: Defect detection software a. Data pre-processing algorithms b. Deep learning training system for machine vision c. Defect detection algorithms Specifications: Runtime environment: Windows / Linux / ARM Speed: < 200ms @ 4M camera Accuracy: > 95% (10-fold) 3. Deep machine vision software platform: Defect classification software a. Deep learning based vision classification algorithms b. Defect classification algorithms Specifications: Runtime environment: Windows / Linux / ARM Speed: < 500ms @ 4M camera with < 50 subjects Accuracy: > 95% (10-fold) 4. 2.5/3D cover glass inspection system a. High stability glass conveyer module for inspection b. Curved glass illumination module c. Multi-view image capture module d. Deep machine vision software system integration & optimization for 2.5D/3D cover glass inspection system e. Prototype pilot-run at cooperation party factory Specifications: Glass size: 3-8 inch Capturing views: front-view, side-view Defect size: > 0.015mm Speed: < 5s (2.5D), <8s (3D) 5. (A) Contract service deliverables for BICI: Deep learning vision classification training system Specification: System accuracy: 90% TPR (True Positive Rate) @ 5% FPR (False Positive Rate) (B) Contract service deliverables for SAE: Customize a smart machine vision software for the Customer’s wafer/chip surface defect inspection Specifications: I Smart machine vision software (alpha version), comprising of: a) Positioning module b) Detection module (alpha version) c) Point-line module d) Calibration module e) Image IO (input-output) module f) System logging module g) Communication module II Smart machine vision software (beta version), comprising of: a) All features in item 1 b) Camera capture module & extending interface c) Detection module (beta version) d) Image pre-processing module e) Plugin-module III Smart machine vision software (final version), comprising of: a) All features in item 2 b) GUI (Graphical User Interface) defect model design module c) Customized defect detection model d) Defect detection module IX A standard operation procedure in PowerPoint slides format (C) Contract service deliverables for Prevision: Design and prototype of dual line 2.5D & 3D cover glass inspection system

    BICI HONGKONG CO., LIMITED
    NEC Hong Kong Limited
    Prevision Technology Limited
    SAE Magnetics (HK) Limited
    ShenZhen Innomaxim Technology Co. Ltd.


    Machine vision technology has been successfully applied into various automated production lines in developed countries in the past years, and recently replaces lots of operators in the assembly production lines of China. Currently, the worldwide manufacturers are encountering the new problem: the increasing cost of the human labor for cosmetic defects inspection is conflicting with the increasing quality demand from the end consumers. The manufactures must spend huge amount of money on hiring workers to perform cosmetic defect inspection. Machine Vision is the most feasible direction for solving this problem, but most of the existing vision technologies are usually only good at dimensional measurement. Some commercial software providers try to build in cosmetic defect inspection modules, which are either template matching based or inflexible coding approaches. Their software cannot fulfill the high requirement of cosmetic defects inspection due to low robustness, long trial run time, and high knowledge requirement in certain domain. Therefore, this project proposes deep machine vision platform technologies to empower the vision application for defect inspection. Along these lines, the project team propose the following IP protected platform technologies to tap the market requirement, including: (1) rapid vision application development technologies for vision measurement software; (2) deep learning defect detection technologies which discover the defect detection rules from a set of vision training data, rather than manually designed by the vision engineers; (3) deep learning based defect classification technologies for advanced defect analysis; (4) 2.5D/3D glass inspection illumination and capture technologies. The proposed technologies can be widely used in different industries, such as consumer electronics (CE) manufacturing, semi-conductor manufacturing, etc. In particular, for CE industries, the major manufacturers locate in China and the revenue is increasing rapidly by years. Taking the mobile phone display manufacturing as example, more than 81% of the products are made in China. The proposed platform technologies can help manufacturers to build up vision solution for inspection purpose in a quick and low cost way. Furthermore, the deep machine vision platform technologies can help the users save 50% of vision application trial run time and reduce 30% of total cost.