Semi-supervised Hierarchical Clustering for Semantic SAR Image Annotation


During this paper, we tend to propose a semi-automated hierarchical clustering and classification framework for artificial aperture radar (SAR) image annotation. Our implementation of the framework allows the classification and annotation of image knowledge starting from scenes up to large satellite knowledge archives. Our framework contains 3 stages: one) every image is cut into patches and every patch is reworked into a texture feature vector; a pair of) similar feature vectors are grouped into clusters, where the quantity of clusters is determined by repeated cluster splitting to optimize their Gaussianity; and 3) the foremost appropriate class (i.e., a semantic label) is assigned to each image patch. This is accomplished by semi-supervised learning. For the testing and validation of our implemented framework, an idea for a two-level hierarchical semantic image content annotation was designed and applied to a manually annotated reference dataset consisting of numerous TerraSAR-X image patches with meter-scale resolution. Here, the higher level contains general classes, whereas the lower level provides additional detailed subclasses for every parent class. For a quantitative and visual evaluation of the proposed framework, we compared the relationships among the clustering results, the semi-supervised classification results, and also the two-level annotations. It turned out that our proposed technique is ready to obtain reliable results for the higher-level (i.e., general class) semantic categories; but, thanks to the too several detailed subclasses versus the few instances of each subclass, the proposed method generates inferior results for the lower level. The foremost important contributions of this paper are the combination of changed Gaussian-means that and modified cluster-then-label algorithms, for the purpose of huge-scale SAR image annotation, yet as the measurement of the clustering and classification performances of various distance metrics.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here

PROJECT TITLE :Unified Discriminative and Coherent Semi-Supervised Subspace Clustering - 2018ABSTRACT:The ubiquitous large, complex, and high dimensional datasets in computer vision and machine learning generates the matter of
PROJECT TITLE :Semi-Supervised Deep Learning Using Pseudo Labels for Hyperspectral Image Classification - 2018ABSTRACT:Deep learning has gained popularity in an exceedingly variety of computer vision tasks. Recently, it's also
PROJECT TITLE :Semi-Supervised Image-To-Video Adaptation For Video Action Recognition - 2017ABSTRACT:Human action recognition has been well explored in applications of pc vision. Many successful action recognition methods have
PROJECT TITLE : Adaptive ensembling of semi-supervised clustering solutions - 2017 ABSTRACT: Conventional semi-supervised clustering approaches have many shortcomings, such as (one) not absolutely utilizing all useful must-link
PROJECT TITLE :Driver Distraction Detection Using Semi-Supervised Machine LearningABSTRACT:Real-time driver distraction detection is that the core to many distraction countermeasures and fundamental for constructing a driver-focused

Ready to Complete Your Academic MTech Project Work In Affordable Price ?

Project Enquiry