Machine Learning Approach Towards Satellite Image Classification

Ferdous, Humayra; Siraj, Tasnim; Setu, Shifat Jahan; Anwar, Musfique; Rahman, Muhammad Arifur

doi:10.1007/978-981-33-4673-4_51

Cited by 21 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At this point, a particular comment can be made regarding the selection of the CNN technique as the enabling method to be integrated with the CI physical model to achieve classification. In this context, other algorithms, such as K-Nearest Neighbors (kNN) [76], [77] and Random Forest [78], [79] techniques, can also be considered ideal for the multi-class classification problem. To this end, the mmW CI classification problem in this paper was also solved using these two algorithms, and the overall classification f1-score accuracies are reported to be 0.86 for the Random Forest algorithm and 0.83 for the kNN algorithm.…”

Section: A Classification Resultsmentioning

confidence: 99%

Coded-Aperture Computational Millimeter-Wave Image Classifier Using Convolutional Neural Network

et al. 2021

View full text Add to dashboard Cite

A millimeter-wave (mmW) classifier system applied to images synthesized from a codedaperture based computational imaging (CI) radar is presented. A developed physical model of a CI system is used to generate the image dataset for the classification algorithm. A convolutional neural network (CNN) is integrated with the physical model and trained using the dataset comprising of synthesized mmW images obtained directly from the developed CI physical model. A k-fold cross validation technique is applied during the training process to validate the classification model. The coded-aperture CI concept enables image reconstruction from a significantly reduced number of back-scattered measurements by facilitating physical layer compression. This physical layer compression can substantially simplify the data acquisition layer of imaging radars, which is realized using only two channels in this article. The integration of the classification algorithm with the CI numerical model is particularly important in enabling the training step to be carried out using relevant system metrics and without the necessity for experimental data. Leveraging the CI numerical model generated data, training step for the classification algorithm is achieved in real-time while also confirming that the numerically trained CI classifier offers high accuracy with both simulated and experimental data. The classifier integrated physical model also enables performance analysis of the classification algorithm to be carried out as a function of key system metrics such as signal-to-noise (SNR) level, ensuring a complete understanding of the classification accuracy under different operating conditions. The trained CI system is tested with synthesized mmW images from the physical model and a classification accuracy of 89% is achieved. The proposed model is also verified using experimental data validating the fidelity of the developed CI integrated classifier system. A classification latency of 3.8 ms per frame is achieved, paving the way for real-time automated threat detection (ATD) for security-screening applications.INDEX TERMS Millimetre-wave, imaging radars, computational imaging, neural networks, image classification, coded-aperture.

show abstract

Section: A Classification Resultsmentioning

confidence: 99%

Coded-Aperture Computational Millimeter-Wave Image Classifier Using Convolutional Neural Network

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The decision tree learning technique is a kind of supervised learning. It is most effective in situations involving instances represented by attribute-value pairs, with the goal function having discrete output values [16]. Instances are categorized using the Decision Tree method, sorting them in ascending order from the tree's root node to its leaf nodes.…”

Section: B Classification Modulementioning

confidence: 99%

A Systematic Video Indexing Approach Using Decision Tree

Madheswaran¹,

Murugaiyan²

2022

IJASIS

View full text Add to dashboard Cite

A systematic categorization approach for video indexing is presented in this paper. The amount of multimedia information that can be accessed over the internet continues to expand exponentially. Due to this growth and development of multimedia on the internet, particularly videos, there has been a rise in the need for video retrieval. The goal of this work is to discover subsets of characteristics that are suitable for indexing or categorizing video content. The features chosen from the frames of the video have dominant texture characteristics. Several statistical features have been applied for better performance with decision tree classification. The investigation included 1000 videos from different video contents, such as news, cartoon, advertisements, movies, and sports categories. Results show that the overall misclassification rate percentage was below 3%. The capability of indexing the video contents indicates the real power of the proposed system, which can enhance existing indexing services, thereby enriching the tools that are available for video indexing.

show abstract

“…There is a vast array of works of machine learning in diverse area i.e. Classification of Fake News [4], Facial Spoof Detection [5], Image classification [6], Auditory attention state [7], Computational biology [8], Trust management for IOT [9], Text processing [10,11] are going.…”

Section: Literature Reviewmentioning

confidence: 99%

A Recommender System for Adaptive Examination Preparation using Pearson Correlation Collaborative Filtering

Hossain,

Tasnim,

Hoque

et al. 2021

Int J Auto AI Mach learn

View full text Add to dashboard Cite

Distance learning is any type of far-off instruction where the understudy isn't actually present for the exercise. It is blasting gratitude to the force of the Internet. Distance learning plays a vital role for examination preparation where multiple choice questions can be utilized to evaluate the performance of students. Multiple Choice Question (MCQ) is a type of question used in the examination to evaluate the performance of students accordingly where usually four options are given along with the question, and one has to choose the correct answer. This research includes a simulation model that has been built to keep the learners continue to learn the subjects they might be weak in. We have developed a methodology that may guide a student to update his/her area of weakness by using a recommender system based on Pearson Correlation Collaborative Filtering approach. The paper describes a recommender system that will keep track of a learner's profile and create an adaptive training mechanism using the performance matrix.

show abstract

Machine Learning Approach Towards Satellite Image Classification

Cited by 21 publications

References 16 publications

Coded-Aperture Computational Millimeter-Wave Image Classifier Using Convolutional Neural Network

Coded-Aperture Computational Millimeter-Wave Image Classifier Using Convolutional Neural Network

A Systematic Video Indexing Approach Using Decision Tree

A Recommender System for Adaptive Examination Preparation using Pearson Correlation Collaborative Filtering

Contact Info

Product

Resources

About