2. Professors
  3. Teaching

Image-based Measurements

Paolo SOMMELLA Image-based Measurements

0622400013
DIPARTIMENTO DI INGEGNERIA INDUSTRIALE
EQF7
ELECTRONIC ENGINEERING
2017/2018

YEAR OF COURSE 2
YEAR OF DIDACTIC SYSTEM 2016
SECONDO SEMESTRE
CFUHOURSACTIVITY
660LESSONS
PAOLO SOMMELLA T Curriculum
Objectives
Learning objectives: expected learning outcomes and competence to be acquired
The course aims to treat topics of major interest related to measurement systems and methods based on the analysis of digital images adopted for the automation of industrial processes. On-line and contact-less systems and methods for vision-based measurements (based on image processing) of dimensional parameters of parts and products will be described, and models related to the evaluation of measurement uncertainty of the results provided will be introduced.

Knowledge and understanding skills
Understanding the terminology used in the development of image processing technicques and contact-less measurement systems.

Knowledge and understanding skills applied
- Designing hardware sub-system for image acquisition and illumination
- Designing and implementation of camera calibration and image-based measurement algorithms;
- Metrological characterization of image-based measurement systems.

Judgment autonomy
Know how to find the most appropriate methods for efficiently designing and implementing a contact-less measurement system for industrial automation.

Communicative Skills
Know how to expose orally a topic related to Image processing and Artificial Vision-based Measurement Systems.
Know how to work in a group to design and realize a measurement system for industrial automation.

Ability to learn
Know how to apply the acquired knowledge to contexts different from those presented during the course. Learn how to decipher the topics dealt with using materials other than those proposed.
Prerequisites
Compulsory pre-requisites are not mandatory but the student should have previously acquired the basics on:
- Linear Algebra and Geometry;
- Computer Science and Programming Language;
- use of Measurement Instruments.
Contents
- Introduction (2 h): Introduction to measurement systems based on image analysis.

- Devices (8 h): Cameras for industrial applications: architecture and features. Introduction to optics. Thin lens model and relationships for the design of optics. The depth of field and the diffraction. Illuminators and their main characteristics. Types of light sources for industrial inspection and criteria for an optimal choice.

- Analysis of digital images (20 h). The digital images. Software architectures for image-based measurements. Point transformations: histogram and its applications, thresholding, contrast enhancement. Local transformations: linear and nonlinear types. Edge detection. Canny’s algorithm. Hough transform.

- Camera model and calibration (10 h). The process of image formation. The "pin hole" model. The equations for a perspective projection. The distortion of the lens. Direct Linear Transformation algorithm (DLT) for camera calibration. The calibration target. Decomposition of the perspective projection matrix. Correction of lens distortion. Propagation of uncertainty through the DLT calibration algorithm.

- Stereo measurement systems (5 h): Measurement systems based on stereo vision. Measurement of the depth in the case of simplifying assumptions, and evaluation of its uncertainty. Stereo reconstruction in the general case with linear methods, and uncertainty propagation. The epipolar geometry.

- Final projects (15 h): develop a measurement application based on the analysis of digital images and on camera calibration. The development of software component of the project will be done in a programming language to be agreed (typically in LabVIEW or C / C + + using OpenCV libraries, but other choices are possible as well).
Teaching Methods
Teaching includes:
- lectures (30 hours),
- numerical exercises (15 hours)
- team project developed in the laboratory (15 hours).
Verification of learning
The achievement of the objectives of the teaching is certified by passing the evaluation exam (vote expressed on a minimum point of 18 and maximum point of 30 cum laude), which provides for :
- a 1st numerical test about the application of image processing tecniques;
- a 2nd numerical test about the application of camera calibration tecniques;
- a final oral exam of an average duration of 30 minutes, aimed at verifying: 1) the learning of the subjects covered in frontal teaching hours; 2) the group project realized on a subject assigned by the teacher.
Texts
- R.C. Gonzalez, R. C. Woods, "Digital Image Processing", 2nd Ed., 2002, Prentice Hall, Inc
- R. Hartley, A. Zisserman "Multiple View Geometry in Computer Vision", 2nd Ed., Cambridge University Press.
  BETA VERSION Data source ESSE3 [Ultima Sincronizzazione: 2019-05-14]