Computer Vision and Image Recognition
Expert-defined terms from the Professional Certificate in AI for Real Estate Professionals course at Stanmore School of Business. Free to read, free to share, paired with a professional course.
Computer Vision and Image Recognition Glossary #
Computer Vision and Image Recognition Glossary
A #
A
1 #
AI (Artificial Intelligence):
- Explanation: AI refers to the simulation of human intelligence processes by ma… #
- Explanation: AI refers to the simulation of human intelligence processes by machines, including learning, reasoning, and self-correction.
2 #
Algorithm:
- Explanation: An algorithm is a set of rules or instructions designed to perfor… #
- Explanation: An algorithm is a set of rules or instructions designed to perform a specific task or solve a problem.
3 #
Annotation:
- Explanation: Annotation involves marking up specific features or objects withi… #
- Explanation: Annotation involves marking up specific features or objects within an image to provide training data for computer vision models.
4 #
Augmented Reality (AR):
- Explanation: AR is a technology that overlays digital information or objects o… #
- Explanation: AR is a technology that overlays digital information or objects onto the real world through a device like a smartphone or headset.
B #
B
5 #
Binary Classification:
- Explanation: Binary classification is the task of categorizing data into one o… #
- Explanation: Binary classification is the task of categorizing data into one of two classes or categories.
6 #
Biometrics:
- Explanation: Biometrics refers to the measurement and analysis of unique physi… #
- Explanation: Biometrics refers to the measurement and analysis of unique physical or behavioral characteristics for identification purposes.
7 #
Computer Vision:
- Explanation: Computer vision is a field of AI that enables machines to interpr… #
- Explanation: Computer vision is a field of AI that enables machines to interpret and understand visual information from the real world.
8 #
Convolutional Neural Network (CNN):
- Explanation: A CNN is a type of neural network designed for processing grid-li… #
- Explanation: A CNN is a type of neural network designed for processing grid-like data, such as images.
C #
C
9 #
Classification:
- Explanation: Classification is the task of assigning labels or categories to i… #
- Explanation: Classification is the task of assigning labels or categories to input data based on its features.
10 #
Deep Learning:
- Explanation: Deep learning is a subset of machine learning that uses neural ne… #
- Explanation: Deep learning is a subset of machine learning that uses neural networks with multiple layers to learn complex patterns in data.
11 #
Dataset:
- Explanation: A dataset is a collection of data used for training and evaluatin… #
- Explanation: A dataset is a collection of data used for training and evaluating machine learning models.
12 #
Feature Extraction:
- Explanation: Feature extraction involves selecting or transforming relevant fe… #
- Explanation: Feature extraction involves selecting or transforming relevant features from raw data to facilitate model training and prediction.
D #
D
13 #
Deep Neural Networks:
14 #
Dimensionality Reduction:
- Explanation: Dimensionality reduction is the process of reducing the number of… #
- Explanation: Dimensionality reduction is the process of reducing the number of features in a dataset while preserving important information.
15 #
Object Detection:
- Explanation: Object detection is the task of identifying and locating objects… #
- Explanation: Object detection is the task of identifying and locating objects within an image or video.
16 #
Preprocessing:
- Explanation: Preprocessing involves preparing and cleaning data before feeding… #
- Explanation: Preprocessing involves preparing and cleaning data before feeding it into a machine learning model.
E #
E
17 #
Edge Detection:
- Explanation: Edge detection is a technique used to identify and highlight edge… #
- Explanation: Edge detection is a technique used to identify and highlight edges or boundaries within an image.
18 #
Enhancement:
- Explanation: Enhancement techniques are used to improve the quality or visual… #
- Explanation: Enhancement techniques are used to improve the quality or visual appearance of an image by adjusting its attributes.
19 #
Evaluation Metrics:
- Explanation: Evaluation metrics are used to measure the performance and effect… #
- Explanation: Evaluation metrics are used to measure the performance and effectiveness of machine learning models.
20 #
Face Recognition:
- Explanation: Face recognition is the task of identifying or verifying individu… #
- Explanation: Face recognition is the task of identifying or verifying individuals based on their facial features.
F #
F
21 #
Feature Engineering:
- Explanation: Feature engineering involves creating new features or transformin… #
- Explanation: Feature engineering involves creating new features or transforming existing features to improve model performance.
22 #
Feature Selection:
- Explanation: Feature selection is the process of choosing the most relevant fe… #
- Explanation: Feature selection is the process of choosing the most relevant features from a dataset to improve model accuracy and efficiency.
23 #
Filter:
- Explanation: A filter is a matrix used in convolutional neural networks to ext… #
- Explanation: A filter is a matrix used in convolutional neural networks to extract features from input data.
24 #
Foreground Extraction:
- Explanation: Foreground extraction is the process of separating the main objec… #
- Explanation: Foreground extraction is the process of separating the main objects or subjects in an image from the background.
G #
G
25 #
Generative Adversarial Network (GAN):
- Explanation: GANs are a type of neural network architecture that consists of t… #
- Explanation: GANs are a type of neural network architecture that consists of two networks (generator and discriminator) trained in a competitive manner.
26 #
Geometric Transformation:
- Explanation: Geometric transformations are used to modify the spatial orientat… #
- Explanation: Geometric transformations are used to modify the spatial orientation or size of objects in an image.
27 #
Gesture Recognition:
- Explanation: Gesture recognition is the process of interpreting human gestures… #
- Explanation: Gesture recognition is the process of interpreting human gestures or movements for interaction with electronic devices.
28 #
GPU (Graphics Processing Unit):
- Explanation: GPUs are specialized processors designed to handle complex graphi… #
- Explanation: GPUs are specialized processors designed to handle complex graphics and parallel computations, commonly used in deep learning tasks.
H #
H
29 #
Haar Cascade:
- Explanation: Haar cascades are classifiers used to detect objects in images by… #
- Explanation: Haar cascades are classifiers used to detect objects in images by analyzing patterns of intensity changes.
30 #
Handwritten Text Recognition:
- Explanation: Handwritten text recognition is the process of converting handwri… #
- Explanation: Handwritten text recognition is the process of converting handwritten text into digital text using machine learning algorithms.
31 #
Histogram Equalization:
- Explanation: Histogram equalization is a technique used to improve the contras… #
- Explanation: Histogram equalization is a technique used to improve the contrast and brightness of an image by adjusting the distribution of pixel intensities.
32 #
Human Pose Estimation:
- Explanation: Human pose estimation is the task of identifying and locating key… #
- Explanation: Human pose estimation is the task of identifying and locating key points on a human body to infer its pose or movements.
I #
I
33 #
Image Classification:
- Explanation: Image classification is the process of categorizing images into p… #
- Explanation: Image classification is the process of categorizing images into predefined classes or categories based on their visual content.
34 #
Image Enhancement:
- Explanation: Image enhancement techniques are used to improve the quality or v… #
- Explanation: Image enhancement techniques are used to improve the quality or visual appearance of an image by adjusting its attributes.
35 #
Image Processing:
- Explanation: Image processing is the analysis and manipulation of digital imag… #
- Explanation: Image processing is the analysis and manipulation of digital images to extract useful information or enhance visual quality.
36 #
Image Recognition:
- Explanation: Image recognition is the ability of a computer to identify and in… #
- Explanation: Image recognition is the ability of a computer to identify and interpret visual information from images or videos.
37 #
Image Segmentation:
- Explanation: Image segmentation is the process of partitioning an image into m… #
- Explanation: Image segmentation is the process of partitioning an image into multiple segments or regions based on certain criteria.
38 #
Image Stitching:
- Explanation: Image stitching is the process of combining multiple images with… #
- Explanation: Image stitching is the process of combining multiple images with overlapping areas to create a wider or panoramic image.
39 #
Instance Segmentation:
- Explanation: Instance segmentation aims to detect and segment individual insta… #
- Explanation: Instance segmentation aims to detect and segment individual instances of objects within an image.
40 #
Interpolation:
- Explanation: Interpolation is a method used to estimate unknown values between… #
- Explanation: Interpolation is a method used to estimate unknown values between known data points to achieve a smooth transition.
J #
J
41 #
Joint Detection:
- Explanation: Joint detection involves identifying and localizing key points or… #
- Explanation: Joint detection involves identifying and localizing key points or joints on an object or human body.
K #
K
42 #
Keypoints:
- Explanation: Keypoints are specific points or regions in an image that are dis… #
- Explanation: Keypoints are specific points or regions in an image that are distinctive and can be used for feature matching or recognition.
43 #
Kernel:
- Explanation: A kernel is a matrix used in convolutional neural networks to per… #
- Explanation: A kernel is a matrix used in convolutional neural networks to perform operations like filtering or feature extraction on input data.
44. K #
Means Clustering:
- Explanation: K-means clustering is a clustering algorithm that partitions data… #
- Explanation: K-means clustering is a clustering algorithm that partitions data into K clusters based on feature similarities.
L #
L
45 #
Labeling:
- Explanation: Labeling involves assigning labels or categories to data points t… #
- Explanation: Labeling involves assigning labels or categories to data points to provide supervision for machine learning models.
46 #
Localization:
- Explanation: Localization is the task of predicting the location or coordinate… #
- Explanation: Localization is the task of predicting the location or coordinates of objects within an image.
47 #
Loss Function:
- Explanation: A loss function is used to measure the error or discrepancy betwe… #
- Explanation: A loss function is used to measure the error or discrepancy between predicted and actual values in a machine learning model.
M #
M
48. Mask R #
CNN:
- Explanation: Mask R-CNN is a neural network architecture that extends Faster R… #
- Explanation: Mask R-CNN is a neural network architecture that extends Faster R-CNN for object detection to also include instance segmentation.
49 #
Matching:
- Explanation: Matching involves finding correspondences or similarities between… #
- Explanation: Matching involves finding correspondences or similarities between features in different images for tasks like object recognition or tracking.
50 #
Metric Learning:
- Explanation: Metric learning is a technique used to learn a distance metric th… #
- Explanation: Metric learning is a technique used to learn a distance metric that optimally represents the relationships between data points.
51 #
Morphological Operations:
- Explanation: Morphological operations are image processing techniques that mod… #
- Explanation: Morphological operations are image processing techniques that modify the shape or structure of objects in an image.
N #
N
52 #
Neural Network:
- Explanation: A neural network is a computational model inspired by the human b… #
- Explanation: A neural network is a computational model inspired by the human brain that consists of interconnected nodes (neurons) organized in layers.
53 #
Normalization:
- Explanation: Normalization is the process of rescaling data to have a mean of… #
- Explanation: Normalization is the process of rescaling data to have a mean of 0 and a standard deviation of 1 to improve model performance.
54 #
Object Detection:
- Explanation: Object detection is the task of identifying and locating objects… #
- Explanation: Object detection is the task of identifying and locating objects within an image or video.
55 #
Object Recognition:
- Explanation: Object recognition is the process of identifying and categorizing… #
- Explanation: Object recognition is the process of identifying and categorizing objects or patterns within an image.
56 #
OCR (Optical Character Recognition):
- Explanation: OCR is the technology used to convert printed or handwritten text… #
- Explanation: OCR is the technology used to convert printed or handwritten text into digital text for analysis or processing.
57 #
Overfitting:
- Explanation: Overfitting occurs when a machine learning model performs well on… #
- Explanation: Overfitting occurs when a machine learning model performs well on training data but poorly on unseen data due to capturing noise or irrelevant patterns.
P #
P
58 #
PCA (Principal Component Analysis):
- Explanation: PCA is a technique used to reduce the dimensionality of data by t… #
- Explanation: PCA is a technique used to reduce the dimensionality of data by transforming it into a set of orthogonal components that capture the most variance.
59. Pixel #
wise Classification:
- Explanation: Pixel-wise classification involves assigning a class label to eac… #
- Explanation: Pixel-wise classification involves assigning a class label to each pixel in an image based on its visual characteristics.
60 #
Pooling:
- Explanation: Pooling is a down-sampling operation used in convolutional neural… #
- Explanation: Pooling is a down-sampling operation used in convolutional neural networks to reduce spatial dimensions and extract dominant features.
61 #
Pretrained Model:
- Explanation: A pretrained model is a machine learning model that has been trai… #
- Explanation: A pretrained model is a machine learning model that has been trained on a large dataset and can be used as a starting point for new tasks.
R #
R
62 #
Random Forest:
- Explanation: Random Forest is an ensemble learning method that builds multiple… #
- Explanation: Random Forest is an ensemble learning method that builds multiple decision trees and combines their predictions for improved accuracy.
63 #
Ranking:
- Explanation: Ranking involves ordering items or results based on their relevan… #
- Explanation: Ranking involves ordering items or results based on their relevance to a specific query or criteria.
64 #
Regression:
- Explanation: Regression is a machine learning task that involves predicting co… #
- Explanation: Regression is a machine learning task that involves predicting continuous outcomes or values based on input features.
S #
S
65 #
Saliency Detection:
- Explanation: Saliency detection is the process of identifying the most visuall… #
- Explanation: Saliency detection is the process of identifying the most visually important or distinctive regions within an image.
66 #
Semantic Segmentation:
- Explanation: Semantic segmentation is the task of assigning semantic labels to… #
- Explanation: Semantic segmentation is the task of assigning semantic labels to each pixel in an image to segment objects or regions.
67 #
Sequence Prediction:
- Explanation: Sequence prediction involves forecasting future values or events… #
- Explanation: Sequence prediction involves forecasting future values or events based on a sequence of input data.
68 #
Siamese Network:
- Explanation: A Siamese network is a neural network architecture that is used t… #
- Explanation: A Siamese network is a neural network architecture that is used to learn similarity between inputs by comparing their representations.
69 #
Simultaneous Localization and Mapping (SLAM):
- Explanation: SLAM is a technique used in robotics to create maps of an environ… #
- Explanation: SLAM is a technique used in robotics to create maps of an environment while simultaneously localizing the robot within it.
70 #
Supervised Learning:
- Explanation: Supervised learning is a machine learning paradigm where models a… #
- Explanation: Supervised learning is a machine learning paradigm where models are trained on labeled data to make predictions on new, unseen data.
71 #
Support Vector Machine (SVM):
- Explanation: SVM is a supervised learning algorithm used for classification ta… #
- Explanation: SVM is a supervised learning algorithm used for classification tasks that finds an optimal hyperplane to separate different classes.
T #
T
72 #
Template Matching:
- Explanation: Template matching involves comparing a template (pattern) with a… #
- Explanation: Template matching involves comparing a template (pattern) with a target image to find the best matching region.
73 #
Text Detection:
- Explanation: Text detection is the process of locating and identifying text wi… #
- Explanation: Text detection is the process of locating and identifying text within images or scenes for further analysis or processing.
74 #
Texture Analysis:
- Explanation: Texture analysis involves extracting and quantifying patterns or… #
- Explanation: Texture analysis involves extracting and quantifying patterns or structures within an image to characterize its texture.
75 #
Transfer Learning:
- Explanation: Transfer learning is a machine learning technique where knowledge… #
- Explanation: Transfer learning is a machine learning technique where knowledge gained from one task is applied to a different but related task.
76 #
Unsupervised Learning:
- Explanation: Unsupervised learning is a machine learning paradigm where models… #
- Explanation: Unsupervised learning is a machine learning paradigm where models learn patterns or structures in data without explicit labels.
77 #
Upsampling:
- Explanation: Upsampling is a technique used to increase the resolution or size… #
- Explanation: Upsampling is a technique used to increase the resolution or size of an image by adding new pixels between existing ones.
V #
V
78 #
Vanishing Gradient Problem:
- Explanation: The vanishing gradient problem occurs in deep neural networks whe… #
- Explanation: The vanishing gradient problem occurs in deep neural networks when