Buch, Englisch, 331 Seiten, Format (B × H): 155 mm x 235 mm, Gewicht: 534 g
Reihe: Maker Innovations Series
Build and Design Vision products Using Python and OpenCV
Buch, Englisch, 331 Seiten, Format (B × H): 155 mm x 235 mm, Gewicht: 534 g
Reihe: Maker Innovations Series
ISBN: 979-8-8688-0096-2
Verlag: Apress
Today’s industries are faced with a growing demand for vision systems due to their non-invasive characteristics in inspecting product quality. These systems identify surface defects and faults, and verify components’ orientation and their measurements, etc. This book explores the vision techniques needed to design and develop your own industrial vision system with the help of Raspberry Pi.
You’ll start by reviewing basic concepts and applications of machine vision systems, followed by the preliminaries of Python, OpenCV, required libraries, and installing OpenCV for Python on Raspberry Pi. These are used when implementing image processing for the system applications. You’ll then look at interfacing techniques and some of the challenges industrial vision systems encounter, such as lighting and camera angles.
Algorithms and image processing techniques are also discussed, along with machine learning and deep learning techniques. Later chapters explain the use of GUI apps and real-time applications of Industrial vision systems. Each chapter concludes with examples and demo implementations to facilitate your knowledge of the concepts.
By the end of the book, you’ll be able to build and deploy computer vision applications with Python, OpenCV, and Raspberry Pi.
What You'll Learn
- Build and deploy industrial vision system using Raspberry Pi and Python programming
- Explore computer vision techniques using Raspberry Pi and OpenCV
- Implement popular vision techniques for industrial applications in real time
- Review modern image processing techniques such as image segmentation, thresholding, and contours
Who This Book Is For
Raspberry Pi and Python enthusiasts interested in computer vision applications; educators, industrialists, and industrial solution providers who want to design vision-based testing products with the help of Raspberry PiZielgruppe
Popular/general
Autoren/Hrsg.
Fachgebiete
Weitere Infos & Material
Industrial Vision Systems with Raspberry Pi
Chapter 1: Introduction to Industrial Vision Systems
Chapter Goal: To describe recent trends and requirements of vision systems in industries
Chapter 2: Raspberry Pi and required software
Chapter Goal: To present the programming techniques in Raspberry Pi board
1. Raspberry Pi boards and pin details
2. Raspbian OS installation
3. Open CV installation on Raspberry Pi
4. Remote access with Raspberry Pi
5. How to use a Raspberry Pi as a Vision Product?
5.1 Interfacing a camera with Raspberry Pi
5.2 Infinite loop
5.3 . Auto start
Chapter 3: OpenCV - Python
Chapter Goal: To practice the required packages such as OpenCV, PIL and Matplotlib for image analysis and visualization.
1. IDEs for python in Raspberry Pi
2. Python programming in Raspberry Pi
2.1. Numpy arrays
2.2. Python Imaging Library(PIL)
5. Visualizing the images using Matplotlib
6. Read and write the images using OpenCV
7. Capturing the video using OpenCV
Chapter 4: Challenges in Industrial Vision Systems
Chapter Goal: To describe about the various challenges when implementing the vision systems in the industry environment.
1. Camera Models
2. Field of View
3. Lighting and illumination
4. Calibration of camera angle and position
5. Frames per second
Chapter 5: Image Processing using OpenCV
Chapter Goal: To discuss about image processing concepts and implement the same using python OpenCV package.
1. Concepts of image acquisition
2. Image and its properties
3. Noise
4. Types of Images
5. Image Enhancement
6. Resizing the images
7 .Edge detection and Filtering using OpenCV
7.1.Mean filter
7.2.Median Filter
7.3.Sobel Filter
7.4.Canny filter
7.5 Case study: Extracting the edges of the gear teeth.
8. Morphological Operations with OpenCV
8.1. Erosion
8.2. Dilation
8.3.Case study: Serial Number extraction using morphological operations
9. Thresholding Techniques with OpenCV
9.1. Binary thresholding
9.2. Binary Inverse thresholding
9.3. Otsu thresholding
9.4. Adaptive thresholding
9.5: Case study: Analysis of thresholding techniques
10. Binary Large Objects(BLOB) detection using OpenCV
10.1. BLOB in images
10.2 Case Study: Missing parts detection in a bearing using Blob
11. Contours detection with OpenCV
12. Measurement of mechanical components using OpenCV
12.1 Resolution and pixel size
12.2 Calibration of Camera for measurements
12.3. Case study: Measuring a length of boring bars in cutting tools
13. Background estimation using OpenCV
Chapter 6: Graphical User Interface with OpenCV
Chapter goal: To develop a customized user interface for vision system applications using OpenCV.
Chapter 7: Feature Detection and Matching
Chapter goal: To describe the features in images and utilize various techniques to extract the features in the images.
1. Features
2. SIFT, FAST, BRIEF, SURF and ORB
3. Haar classifiers
4. Corner detection
5. Shape and Line detection in images
Chapter 8: Image segmentation
Chapter goal: To study the segmentation techniques for extracting the particular object in an image.
Chapter 9: Optical Character Recognition
Chapter goal: To identify the text or numbers in mechanical parts like shaft, gears, bearings, etc.
Chapter 10: Machine learning techniques for vision applications
Chapter Goal: To implement the machine learning techniques for mechanical parts detection, missing part identification, and surface defect inspection.
1. Machine learning models for image data
2. Convolutional Neural Network(CNN)
3. Support Vector Machines(SVM)
4. Case study: Mechanical parts identification using CNN
Chapter 11: Industrial Vision system Applications
Chapter Goal: To apply the vision algorithms for the real time case studies as mentioned below:
1. Shape inspection in keys using contours
2. Chamfer flywheel position detection using template matching
3. Missing part identification using template matching
4. Surface defects inspection using contours and background subtraction
5. Thread count inspection in Nut and Bolt using contours extraction
