Checking date: 10/06/2022


Course: 2022/2023

Machine Vision
(19217)
Study: M. Applied Artificial Intelligence (378)
EPI


Coordinating teacher: ESCALERA HUESO, ARTURO DE LA

Department assigned to the subject: Department of Systems Engineering and Automation

Type: Electives
ECTS Credits: 3.0 ECTS

Course:
Semester:




Objectives
Artificial Vision, also known as Computer Vision, consists of the automatic analysis of images by computers to determine which objects are present in them. It is a technology widely used in industrial environments for quality control and robot guidance thanks to Machine Learning techniques. During the last ten years, Deep Learning has made its range of applications out of the industrial environment and there are currently numerous applications outside of controlled environments both industrial, for the Internet of Things or for mobile phones. It can be assured that it is Artificial Intelligence that has developed this technology. During the course, the main algorithms that are currently being used both in the industrial field and outside it will be described, with special emphasis on deep learning and with a practical approach.
Skills and learning outcomes
Description of contents: programme
1.- Introduction to Computer Vision. 1.1. Definitions. 1.2. History 1.3. Modules 1.4. Human vision sense 1.5. Applications 2.- Digital images. 2.1. Spatial sampling, grey levels. 2.2. Pixels. 2.3. Arithmetical and logical Operations. 2.4. Colour. 3.- Image Pre-processing. 3.1. Contrast 3.2. Noise reduction 3.3. Image sharpening 3.4 Edge detection. 4.- Segmentation. 4.1. Thresholding and labelling. 4.2. Region growing. 4.3. Split & Merge. 4.4. Mean-Shift 5.- Morphological Transforms and object description. 5.1. Morphological Transforms for binary images 5.2. Morphological Transforms for grey level images 5.3. Region descriptors. 5.4. Shape descriptors. 6.- Object recognition. 6.1. Basic concepts. 6.2. Bayes classifier. 6.3. Clustering. 7. Neural Networks 7.1 Introduction 7.2 Fully-Connected Neural networks 7. 3 Loss function, gradient descent and retro-propagation 8 Convolutional Neural Networks 8.1 From Fully-Connected Layers to Convolutions 8.2 Convolutional Layers 8.3 Multiple Input and Multiple Output Channels 8.4 LeNe 9 Image classifiers 9.1 AlexNet 9.2 VGG 9.3 NiN 9.4 GoogLeNet 9.5 ResNet 9.5 DenseNet 10. Object detection 10.1 Image Augmentation 10.2 Fine-Tuning 10.3 Object Detection and Bounding Boxes 10.4 Multiscale Object Detection 10.5 R-CNNs 10.6 Yolo 10.7 Semantic Segmentation
Learning activities and methodology
Theoretical classes Laboratory practices Tutorials Group work Individual student work Partial exams
Assessment System
  • % end-of-term-examination 0
  • % of continuous assessment (assigments, laboratory, practicals...) 100
Calendar of Continuous assessment
Basic Bibliography
  • Arturo de la Escalera. Visión por computador: fundamentos y métodos. Prentice Hall. 2001
  • Ian Goodfellow and Yoshua Bengio and Aaron Courville. Deep Learning. MIT Press. 2016
Recursos electrónicosElectronic Resources *
(*) Access to some electronic resources may be restricted to members of the university community and require validation through Campus Global. If you try to connect from outside of the University you will need to set up a VPN


The course syllabus may change due academic events or other reasons.