A GRAPH BASED OBJECT DESCRIPTION AND RECOGNITION METHODOLOGY

2008 ◽  
Vol 17 (06) ◽  
pp. 1161-1194 ◽  
Author(s):  
N. BOURBAKIS ◽  
P. YUAN ◽  
P. KAKUMANU
Keyword(s):  
Image Segmentation ◽  
Graph Representation ◽  
Object Model ◽  
Segmentation Method ◽  
Graph Representations ◽  
Object Description ◽  
3D Objects ◽  
Local Graph ◽  
Intermediate Views ◽  
Object Models

This paper presents a methodology for recognizing 3D objects using synthesis of 2D views. In particular, the methodology uses wavelets for rearranging the shape of the perceived 2D view of an object for attaining a desirable size, local-global (LG) graphs for representing the shape, color and location of each image object's region obtained by an image segmentation method and the synthesis of these regions that compose that particular object. The synthesis of the regions is obtained by composing their local graph representations under certain neighborhood criteria. The LG graph representation of the extracted object is compared against a set of LG based object-models stored in a Database (DB). The methodology is accurate for recognizing objects existed in the DB and it has the capability of "learning" the LG patterns of new objects by associating them with attributes from existing LG patterns in the DB. Note that for each object-model stored in the database there are only six views, since all the intermediate views can be generated by appropriately synthesizing these six views. Illustrative examples are also provided.

Multiresolution Bio-Medical Image Segmentation using Fuzzy C-Means Clustering

2019 ◽  
Vol 8 (4) ◽  
pp. 9548-9551
Keyword(s):  
Image Segmentation ◽  
Medical Image Segmentation ◽  
Experimental Result ◽  
Segmentation Method ◽  
Fuzzy C Means ◽  
Fuzzy C Means Clustering ◽  
Robust Image ◽  
Multiple Clusters ◽  
Single Pixel ◽  
Popular Image

Fuzzy c-means clustering is a popular image segmentation technique, in which a single pixel belongs to multiple clusters, with varying degree of membership. The main drawback of this method is it sensitive to noise. This method can be improved by incorporating multiresolution stationary wavelet analysis. In this paper we develop a robust image segmentation method using Fuzzy c-means clustering and wavelet transform. The experimental result shows that the proposed method is more accurate than the Fuzzy c-means clustering.

scholarly journals Hyperspectral Image Classification with Localized Graph Convolutional Filtering

2021 ◽  
Vol 13 (3) ◽  
pp. 526
Author(s):  
Shengliang Pu ◽  
Yuanfeng Wu ◽  
Xu Sun ◽  
Xiaotong Sun
Keyword(s):  
Hyperspectral Image ◽  
Principal Component ◽  
Representation Learning ◽  
Classification Performance ◽  
Hyperspectral Data ◽  
Spectral Graph Theory ◽  
Feature Reduction ◽  
Graph Representation ◽  
Novel Method ◽  
Local Graph

The nascent graph representation learning has shown superiority for resolving graph data. Compared to conventional convolutional neural networks, graph-based deep learning has the advantages of illustrating class boundaries and modeling feature relationships. Faced with hyperspectral image (HSI) classification, the priority problem might be how to convert hyperspectral data into irregular domains from regular grids. In this regard, we present a novel method that performs the localized graph convolutional filtering on HSIs based on spectral graph theory. First, we conducted principal component analysis (PCA) preprocessing to create localized hyperspectral data cubes with unsupervised feature reduction. These feature cubes combined with localized adjacent matrices were fed into the popular graph convolution network in a standard supervised learning paradigm. Finally, we succeeded in analyzing diversified land covers by considering local graph structure with graph convolutional filtering. Experiments on real hyperspectral datasets demonstrated that the presented method offers promising classification performance compared with other popular competitors.

scholarly journals Active Learning with Bayesian UNet for Efficient Semantic Image Segmentation

2021 ◽  
Vol 7 (2) ◽  
pp. 37
Author(s):  
Isah Charles Saidu ◽  
Lehel Csató
Keyword(s):  
Neural Network ◽  
Image Segmentation ◽  
Active Learning ◽  
Convolutional Neural Network ◽  
Medical Image ◽  
Segmentation Method ◽  
Semantic Image Segmentation ◽  
Batch Normalization ◽  
Set Up ◽  
Image Datasets

We present a sample-efficient image segmentation method using active learning, we call it Active Bayesian UNet, or AB-UNet. This is a convolutional neural network using batch normalization and max-pool dropout. The Bayesian setup is achieved by exploiting the probabilistic extension of the dropout mechanism, leading to the possibility to use the uncertainty inherently present in the system. We set up our experiments on various medical image datasets and highlight that with a smaller annotation effort our AB-UNet leads to stable training and better generalization. Added to this, we can efficiently choose from an unlabelled dataset.

scholarly journals Panicle-SEG: a robust image segmentation method for rice panicles in the field based on deep learning and superpixel optimization

Plant Methods ◽  
2017 ◽  
Vol 13 (1) ◽  
Author(s):  
Xiong Xiong ◽  
Lingfeng Duan ◽  
Lingbo Liu ◽  
Haifu Tu ◽  
Peng Yang ◽  
...  
Keyword(s):  
Image Segmentation ◽  
Deep Learning ◽  
Segmentation Method ◽  
Robust Image ◽  
Rice Panicles
Sign in / Sign up

Export Citation Format

Share Document