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

**(True or False)** ImageNet is a classification CNN.

(True or False) ImageNet is a classification CNN.

A ```numpy.array``` of shape `(N, 3, H, W)`

A PIL `Image` object of shape `(H, W, 3)`

A ```torch.Tensor``` of shape `(N, 3, H, W)`, after having been transformed by the transformation pipeline supplied with the pretrained weights

A ```torch.Tensor``` of shape `(3, H, W)`

What type of object can be passed as input to the GoogLeNet CNN for inference?

What type of object can be passed as input to the GoogLeNet CNN for inference?

Suppose a mini-batch of 16 images is passed as input to a GoogLeNet CNN, which outputs 1000 numbers. What is the shape of the output tensor?

Suppose a mini-batch of 16 images is passed as input to a GoogLeNet CNN, which outputs 1000 numbers. What is the shape of the output tensor?

When we instantiate a pretrained CNN, it is set by default to training mode. How to set it to inference mode?

When we instantiate a pretrained CNN, it is set by default to training mode. How to set it to inference mode?

They are used as-is, to classify images of objects from the assembly line.

They serve as an inspiration for designing new neural network architectures.

They serve as a starting point for transfer learning or fine-tuning, to classify images of objects from the assembly line.

They are used to detect the presence of an object, which triggers some action.

In the context of automated inspection, what is the main use of pretrained classification CNNs?

In the context of automated inspection, what is the main use of pretrained classification CNNs?

We replace the ```fc``` layer with another ```torch.nn.Linear``` having six output features.

We arbitrarily choose six output features for our classes, and ignore the 994 other features.

We change all the ```torch.nn.Linear``` layers in the CNN to accommodate the shape of our input tensors.

We identify the classes in the ImageNet challenge that look the most like our six classes, and we use their indices as our target output.

If we have a classification task with six classes, and we want to use transfer learning with the pretrained GoogLeNet CNN, what should be done before starting the training?

If we have a classification task with six classes, and we want to use transfer learning with the pretrained GoogLeNet CNN, what should be done before starting the training?

Pretrained Classification CNNs and Classification CNN Training

Pretrained Classification CNNs and Classification CNN Training

Test your knowledge on pretrained classification CNNs and training a classification CNN.

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: Pretrained Classification CNNs and CNN Training