1. 2D EiH flexible inspection platform (a). Develop & design hybrid light field illumination Dark field illumination module: (1). Light diverging angle of collimated light beam: ≤ ±5 degree; (2). Light intensity: ≥ 10000lux; (3). Light uniformity: ≥90%; (4). Width of lighting beam ≥ 10mm; Bright field illumination module: (1). Light intensity: ≥ 5000lux; (2). Light uniformity: ≥ 90%; (3). Effective width of light emitting area: ≥250mm* 20mm (b). Develop & design fast focus image capture Image quality: MTF>0.3 @70lps/mm; Image sensor size ≥ 2/3 inch; Working distance: ≥ 200mm; Depth of focus: ≥ 30mm; Focus re-setting time: ≤ 25ms (c). Develop & design one shot multi-illumination control for image capture Image line numbers with different illumination condition for one shot: ≥ 3; Capture trigger method: hardware trigger; Capture trigger delay: ≤ 10us; Control different illumination modules in sequence: support and the response time ≤ 20us (d). Develop 2D EiH defect inspection algorithms, software & related system integration Specifications: (1). FOV: ≤200mm; (2). Working distance: 200~500mm; (3). Optical depth: >30mm;(4). Resolution: 0.2mm; (5). Test speed: 100mm/s 2. 3D EiH flexible inspection platform (a). Develop & design high resolution 3D EiH illumination & image capturing system Resolution of 3D pattern projection module: ~4K×2K; Tiled angle of projection engine: 30°~45°; Working distance of projection engine: ≥200mm; Camera resolution: ≥ 4M; FOV of camera lens: ≤300mm×225mm (b). Develop & design embedded 3D image processing unit Camera Interface: camera link, USB 3.0; Parallel GPU SoC design ≥2 chips; Connectivity interface: GigE, USB 3.0 (c). Develop 3D EiH measurement & defect inspection algorithms, software & related system integration Specifications: (1). FOV: ≤300mm×225mm; (2). Working distance: 200~500mm; (3). Optical depth: >20mm;(4). 3D accuracy: 0.5mm 3. Contract service deliverables for O-Net Communications (A) Optical isolator surface inspection system: 1 set (B) Design documents & related software SDK of Optical isolator surface inspection system 4. Contract service deliverables for Shenzhen InnoMaxim (A) High precision 3D measurement optical inspection module and design documents (B)Single direction 2D image capture optical system for large size object inspection and design documents (C) 3D illumination and image capture optical system for large size object inspection and design documents (D) 3D illumination and image capture optical system for small size object inspection and design documents (E)Multi-direction 2D image capture optical system for small size object inspection and design documents 5. Contract service deliverables for Info Bright (A1) Ceramic substrate surface defects inspection module : 1 set (A2) Semi-auto image processing based defect detection SDK & operation procedure & user interface software: 1 set (B) (A1) & (A2) design documents & related usual manual for the inspection module (C1) System integration & optimization testing report (C2) Copper plating ceramic substrate & PR pattern ceramic substrate automatic defect detection software
Info Bright Holdings Limited
Info Bright Holdings Limited (CS) [Sponsor]
Info Bright Holdings Limited (TL) [Sponsor]
IView Displays (Hong Kong) Company Limited [Sponsor]
Kopin (HK) Limited [Sponsor]
O-Net Communications (Shenzhen) Limited [Sponsor]
O-Net Communications (Shenzhen) Limited (CS) [Sponsor]
ShenZhen Innomaxim Technology Co. Ltd. [Sponsor]
ShenZhen Innomaxim Technology Co. Ltd. (CS) [Sponsor]
The Hong Kong University of Science and Technology
Industrial robots have been adopted in factories worldwide for more than fifty years. In recent years, due to ever increasing wages, diminishing worker supply and shortened product life in consumer electronics, robots are required to be more intelligent for purpose of cooperation with human and even replacement of some workforce. Now most automatic inspection machines are designed to target testing objects which have flat surfaces and simple shapes, and hence are unable to handle those with multiple and shiny surfaces or irregular shapes. Eye-in-hand machine vision technology enables the enhancement of the flexibility of the inspection system on the manufacturing line. It integrates the optical system (including flexible illumination and image capture) on the robot arm. As the robot arm moves, it captures and processes 2D or 3D images of the testing object from different angles.
This project aims to develop an eye-in-hand flexible visual inspection platform, on solving the problems of inspecting free form surfaces and realizing automatic surface inspection for complex shape products of automotive component and home appliance. To achieve this goal, the project team will develop the following IP protected platform technologies, including (1) 2D EiH flexible inspection platform which bases on hybrid light field illumination and deep learning based background subtraction algorithm (2) 3D EiH flexible inspection platform which bases on high resolution illumination and image capturing system, embedded 3D image processing unit and curve embedding distributional outlier detection algorithm.
The global size of machine vision system market is USD 1.52 billion in 2014 and is estimated to reach USD 5.49 billion in 2019. In China, the potential market of machine vision can even reach RMB 96 billion per year. Hong Kong government has been taking many efforts to promote and support the Innovative and Technology industry, as well as the initiatives of “Hong Kong Re-industrialization” and the “Guangdong-Hong Kong-Macao Big Bay Area (GHMBBA)” The technologies developed in this project will strengthen the competitiveness of Hong Kong and China companies in this rapid growing industrial robot market.