Artificial Intelligence | Class 12 CBSE
Terms & Definitions
- Computer-Vision – A field of AI that enables machines to interpret and process visual information from the environment.
- Image-Processing – Techniques used to enhance, analyze, and transform digital images.
- Pixel-Value – The intensity or color information stored at a specific point in a digital image.
- Grayscale-Image – An image where each pixel carries intensity information, from black (0) to white (255).
- RGB-Model – A color representation using Red, Green, and Blue channels.
- HSV-Model – A color model using Hue, Saturation, and Value for better color segmentation.
- Noise-Reduction – The process of removing random variations in brightness or color in an image.
- Image-Smoothing – Blurring techniques to reduce detail and noise (e.g., Gaussian blur).
- Edge-Detection – Identifying sharp changes in brightness to highlight boundaries in images.
- Sobel-Operator – An edge detection method based on calculating gradients.
- Canny-Edge-Detector – A multi-stage edge detection algorithm widely used in computer vision.
- Thresholding-Technique – Converting an image into binary form (black and white) based on intensity.
- Adaptive-Thresholding – Thresholding where the threshold value changes depending on local regions of the image.
- Morphological-Operations – Image transformations like erosion, dilation, opening, and closing to modify shapes.
- Erosion-Operation – Removes pixels at object boundaries, making shapes smaller.
- Dilation-Operation – Adds pixels to object boundaries, making shapes larger.
- Contour-Detection – Identifying continuous curves that bound objects in an image.
- Feature-Extraction – Identifying unique attributes (edges, textures, shapes) from an image.
- Face-Detection – Identifying and locating human faces in an image or video.
- Object-Detection – Locating and classifying objects within an image.
- YOLO-Algorithm – (“You Only Look Once”) – A real-time object detection algorithm.
- Haar-Cascade – A machine learning-based method for object detection, commonly used for face recognition.
- Optical-Character-Recognition (OCR) – Converting images of text into editable digital text.
- Image-Segmentation – Dividing an image into multiple segments for easier analysis.
- Semantic-Segmentation – Assigning a class label to each pixel in an image.
- Instance-Segmentation – Differentiating individual objects of the same class within an image.
- Convolutional-Neural-Network (CNN) – A deep learning model specialized for image recognition and classification.
- Pooling-Layer – A CNN component that reduces spatial dimensions while retaining important features.
- Feature-Map – The output of convolution layers that highlights detected features in an image.
- Transfer-Learning – Using a pre-trained model (e.g., VGG, ResNet) for solving new image recognition tasks.
