Automated Inspection with Computer Vision.png

AICV_jupyterOpenCV.tar.gz

load_image

annotation

color_spaces

thresholding

smoothing

masking

jup_smooth

morphology

blob_detection

jup_fiducials

edge_detection

jupyter_masking_edge_detection

corner_detection

jupyter_corner_detection

homographies

jup_homographies

topography

jup_topography

tensor_manipulation

neural_network_building_blocks

training_neural_network

jup_train_nn

cnn_building_blocks

cnn_for_classification

pretrained_classification_cnn

jup_train_classification_cnn

transfer_learning

jup_finetune_classification_CNN

cnn_for_object_detection

jup_pretrained_object_detection

cnn_for_semantic_segmentation

jup_pretrained_semantic_segmentation

annotation_formats

spa_train_mlp

prj_jup_create_train_mlp_classification

Computer vision is essential for developers who wish to learn practical skills, e.g., in industrial manufacturing, automated inspection of products is crucial for quality assurance.

In this course, you’ll apply computer vision and machine learning to analyze images for automated inspection. You’ll start by learning image I/O operations, thresholding, smoothing, and masking. You’ll learn feature detection using template matching and morphology. You’ll use the Sobel and Canny Edge Detectors, Harris Corner Detector, and Hough transform. You’ll learn about 2D transformations, including perspective and affine transformation, and 3D-to-2D projections. You’ll perform topography with a laser line to inspect the 3D shape of an object. Lastly, you’ll train convolutional neural networks for object detection and image segmentation and annotate datasets using CVAT.

By the end of this course, you will have the skills to build inspection systems for industrial applications using computer vision and machine learning.

Automated Inspection with Computer Vision

What topographic information is provided by a single image acquired by the camera?

What topographic information is provided by a single image acquired by the camera?

The product of the conveyor speed and the delay between image captures.

The height of the camera above the conveyor divided by the number of pixels in $x$

What determines the resolution along the z-axis i.e., the distance between two topographic slices?

What determines the resolution along the z-axis i.e., the distance between two topographic slices?

To avoid interference of the coherent laser light with the ambient light

To maximize the contrast of the intense laser line and avoid saturation

Why should we set the sensitivity of the camera to a relatively low level?

Why should we set the sensitivity of the camera to a relatively low level?

The calibration object is not wide enough to cover the full image width.

There is a lack of synchronization between the laser and the camera.

Occlusions might prevent certain parts of the object from being visible to the camera.

Occlusions might prevent the laser line from reaching certain parts of the object.

**(Select all that apply.)** Which situation could cause the laser line to be invisible for some $x$’s? $x$ is the horizontal pixel coordinate.

(Select all that apply.) Which situation could cause the laser line to be invisible for some <math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math>x’s? <math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math>x is the horizontal pixel coordinate.

The width of the laser line is too thick.

The maximum image intensity over the column $x$ is below some threshold.

One of the neighbor columns have a $y_{max}$ value that shows a large discontinuity.

What could indicate a bad laser signal at a given $x$?

What could indicate a bad laser signal at a given <math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>x</mi></mrow><annotation encoding="application/x-tex">x</annotation></semantics></math>x?

We need to match features between the two images.

We need at least two linear equations to solve for the two unknowns $a$ and $b$, at each $x$.

We have a backup plan to find $y_{max}$ for a given $x$, in case there is an occlusion in one of the images.

One image is for the topography, and one image is for the color.

To calibrate the system, why do we need at least two images of the laser line hitting flat horizontal surfaces at different heights?

To calibrate the system, why do we need at least two images of the laser line hitting flat horizontal surfaces at different heights?

**(True or False)** During the calibration, we rely on the fact that all the elements of the topography acquisition system are perfectly aligned.

(True or False) During the calibration, we rely on the fact that all the elements of the topography acquisition system are perfectly aligned.

**(True or False)** When solving a system of three linear equations in two unknowns, we must ignore one of the equations.

(True or False) When solving a system of three linear equations in two unknowns, we must ignore one of the equations.

If we want to calibrate our system with a quadratic relationship between $Y$ and $y_{max}$, what is the minimum number of images of the laser line hitting an object at a known height that we must have?

If we want to calibrate our system with a quadratic relationship between <math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><mi>Y</mi></mrow><annotation encoding="application/x-tex">Y</annotation></semantics></math>Y and <math xmlns="http://www.w3.org/1998/Math/MathML"><semantics><mrow><msub><mi>y</mi><mrow><mi>m</mi><mi>a</mi><mi>x</mi></mrow></msub></mrow><annotation encoding="application/x-tex">y_{max}</annotation></semantics></math>ymax​, what is the minimum number of images of the laser line hitting an object at a known height that we must have?

Topography with Laser Line

Topography with Laser Line

Test your knowledge of the topography acquisition system with a laser line.

Introduction

Getting Started with Images

Image I/O and Annotations

Color Spaces and Thresholding

Convert Color Spaces, Threshold

Smoothing and Masking

Detection of Features

Image Registration

3D Vision

Getting Started with Neural Networks

Convolutional Neural Networks

Project: Create and Train a CNN for Classification

Object Detection and Semantic Segmentation

Dataset Annotation

Final Remarks

Quiz: Topography with Laser Line