OpenCV 4 with Python Blueprints: Become proficient in computer vision by designing advanced projects using OpenCV 4 with Python 3.8

Cover
Title Page
Copyright and Credits
About Packt
Contributors
Table of Contents
Preface
Chapter 1: Fun with Filters
	Getting started
	Planning the app
	Creating a black-and-white pencil sketch
		Understanding approaches for using dodging and burning techniques
		Implementing a Gaussian blur with two-dimensional convolution
		Applying pencil sketch transformation
			Using an optimized version of a Gaussian blur
	Generating a warming and cooling filter
		Using color manipulation via curve shifting
		Implementing a curve filter using lookup tables
		Designing the warming and cooling effect
	Cartoonizing an image
		Using a bilateral filter for edge-aware smoothing
		Detecting and emphasizing prominent edges
		Combining colors and outlines to produce a cartoon
	Putting it all together
		Running the app
		Mapping the GUI base class
			Understanding the GUI constructor
				Learning about a basic GUI layout
				Handling video streams
		Drafting a custom filter layout
	Summary
	Attributions
Chapter 2: Hand Gesture Recognition Using a Kinect Depth Sensor
	Getting started
	Planning the app
	Setting up the app
		Accessing the Kinect 3D sensor
		Utilizing OpenNI-compatible sensors
		Running the app and main function routine
	Tracking hand gestures in real time
	Understanding hand region segmentation
		Finding the most prominent depth of the image center region
		Applying morphological closing for smoothening
		Finding connected components in a segmentation mask
	Performing hand shape analysis
		Determining the contour of the segmented hand region
		Finding the convex hull of a contour area
		Finding the convexity defects of a convex hull
	Performing hand gesture recognition
		Distinguishing between different causes of convexity defects
		Classifying hand gestures based on the number of extended fingers
	Summary
Chapter 3: Finding Objects via Feature Matching and Perspective Transforms
	Getting started
	Listing the tasks performed by the app
	Planning the app
	Setting up the app
		Running the app – the main() function routine
		Displaying results
	Understanding the process flow
	Learning feature extraction
		Looking at feature detection
		Detecting features in an image with SURF
		Obtaining feature descriptors with SURF
	Understanding feature matching
		Matching features across images with FLANN
		Testing the ratio for outlier removal
		Visualizing feature matches
		Mapping homography estimation
		Warping the image
	Learning feature tracking
		Understanding early outlier detection and rejection
	Seeing the algorithm in action
	Summary
	Attributions
Chapter 4: 3D Scene Reconstruction Using Structure from Motion
	Getting started
	Planning the app
	Learning about camera calibration
		Understanding the pinhole camera model
		Estimating the intrinsic camera parameters
			Defining the camera calibration GUI
			Initializing the algorithm
			Collecting image and object points
			Finding the camera matrix
	Setting up the app
		Understanding the main routine function
		Implementing the SceneReconstruction3D class
	Estimating the camera motion from a pair of images
		Applying point matching with rich feature descriptors
		Using point matching with optic flow
		Finding the camera matrices
		Applying image rectification
	Reconstructing the scene
	Understanding 3D point cloud visualization
	Learning about structure from motion
	Summary
Chapter 5: Using Computational Photography with OpenCV
	Getting started
	Planning the app
	Understanding the 8-bit problem
		Learning about RAW images
		Using gamma correction
	Understanding high-dynamic-range imaging
		Exploring ways to vary exposure
			Shutter speed
			Aperture
			ISO speed
		Generating HDR images using multiple exposure images
			Extracting exposure strength from images
			Estimating the camera response function
		Writing an HDR script using OpenCV
		Displaying HDR images
	Understanding panorama stitching
		Writing script arguments and filtering images
		Figuring out relative positions and the final picture size
			Finding camera parameters
			Creating the canvas for the panorama
			Blending the images together
		Improving panorama stitching
	Summary
	Further reading
	Attributions
Chapter 6: Tracking Visually Salient Objects
	Getting started
	Understanding visual saliency
	Planning the app
	Setting up the app
		Implementing the main function 
		Understanding the MultiObjectTracker class
	Mapping visual saliency
		Learning about Fourier analysis
		Understanding the natural scene statistics
		Generating a saliency map with the spectral residual approach
		Detecting proto-objects in a scene
	Understanding mean-shift tracking
		Automatically tracking all players on a soccer field
	​Learning about the OpenCV Tracking API 
	Putting it all together
	Summary
	Dataset attribution
Chapter 7: Learning to Recognize Traffic Signs
	Getting started
	Planning the app
	Briefing on supervised learning concepts
		The training procedure
		The testing procedure
	Understanding the GTSRB dataset
		Parsing the dataset
	Learning about dataset feature extraction
		Understanding common preprocessing
		Learning about grayscale features
		Understanding color spaces
		Using SURF descriptor
		Mapping HOG descriptor
	Learning about SVMs
		Using SVMs for multiclass classification
		Training the SVM
		Testing the SVM
			Accuracy
			Confusion matrix
			Precision
			Recall
	Putting it all together
	Improving results with neural networks
	Summary
	Dataset attribution
Chapter 8: Learning to Recognize Facial Emotions
	Getting started
	Planning the app
	Learning about face detection
		Learning about Haar-based cascade classifiers
		Understanding pre-trained cascade classifiers
		Using a pre-trained cascade classifier
		Understanding the FaceDetector class
			Detecting faces in grayscale images
			Preprocessing detected faces
				Detecting the eyes
				Transforming the face
	Collecting data
		Assembling a training dataset
			Running the application
			Implementing the data collector GUI
				Augmenting the basic layout
				Processing the current frame
				Storing the data
	Understanding facial emotion recognition
		Processing the dataset
			Learning about PCA
		Understanding MLPs
			Understanding a perceptron
			Knowing about deep architectures
		Crafting an MLP for facial expression recognition
			Training the MLP
			Testing the MLP
			Running the script
	Putting it all together
	Summary
	Further reading
	Attributions
Chapter 9: Learning to Classify and Localize Objects
	Getting started
	Planning the app
	Preparing an inference script
	Preparing the dataset
		Downloading and parsing the dataset
		Creating a TensorFlow dataset 
	Classifying with CNNs
		Understanding CNNs
		Learning about transfer learning
		Preparing the pet type and breed classifier
		Training and evaluating the classifier
	Localizing with CNNs
		Preparing the model
		Understanding backpropagation
		Training the model
	Seeing inference in action
	Summary
	Dataset attribution
Chapter 10: Learning to Detect and Track Objects
	Getting started
	Planning the app
	Preparing the main script 
		Detecting objects with SSD
		Using other detectors
		Understanding object detectors
			The single-object detector
			The sliding-window approach
			Single-pass detectors
			Learning about Intersection over Union
		Training SSD- and YOLO-like networks 
		Tracking detected objects
	Implementing a Sort tracker
		Understanding the Kalman filter
		Using a box tracker with the Kalman filter
			Converting boundary boxes to observations
			Implementing a Kalman filter
		Associating detections with trackers
		Defining the main class of the tracker
	Seeing the app in action
	Summary
Appendix A: Profiling and  Accelerating Your Apps
	Accelerating with Numba
		Accelerating with the CPU
		Understanding Numba, CUDA, and GPU acceleration
Appendix B: Setting Up a Docker Container
	Defining a Dockerfile
		Working with a GPU
Other Books You May Enjoy
Index