Automated image and video object detection based on hybrid heuristic-based U-net segmentation and faster region-convolutional neural network-enabled learning.
Rajashekar Reddy PalleRavi BodaPublished in: Multim. Tools Appl. (2023)
Keyphrases
- image regions
- object detection
- input image
- grey level
- homogeneous regions
- segmentation method
- image analysis
- segmentation algorithm
- image segmentation
- region segmentation
- multiscale
- segmented images
- boundary information
- optimal segmentation
- test images
- object detectors
- image pixels
- watershed transformation
- convolutional neural network
- image structure
- region of interest
- image segments
- single image
- image data
- scene understanding
- region growing
- segmented regions
- foreground and background
- pixel level
- energy functional
- adjacent regions
- energy function
- image features
- edge detection
- high resolution
- gradient information
- segmentation result
- video data
- object segmentation
- region merging
- seed points
- segmentation scheme
- medical images
- gray value
- image classification
- image representation
- foreground background segmentation
- bounding box