Enhancing Convolutional Neural Network using Hu’s Moments

AbuRass, Sanad; Huneiti, Ammar; Al-Zoubi, Mohammad Belal

doi:10.14569/ijacsa.2020.0111216

Cited by 8 publications

(3 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Where 𝑓(𝑥, 𝑦) is the pixel intensity at (𝑥, 𝑦), and (𝑥̅ , 𝑦 ̅) is the centroid of the image [15], [16].…”

Section: Image Segmentation: Cannymentioning

confidence: 99%

Automated Classification of Empon Plants: A Comparative Study Using Hu Moments and K-NN Algorithm

Hayatou Oumarou,

Rismayanti

2024

ijodas

View full text Add to dashboard Cite

The study "Automated Classification of Empon Plants: A Comparative Study Using Hu Moments and K-NN Algorithm" investigates the potential of image processing and machine learning techniques in the classification of empon plants, specifically ginger and turmeric. Utilizing a dataset of leaf images, the research employed the Canny method for image segmentation and Hu Moments for feature extraction, followed by classification using the K-Nearest Neighbors (K-NN) algorithm. The performance of the model was evaluated through a 5-fold cross-validation method, focusing on metrics such as accuracy, precision, recall, and F1-score. The results showcased the model's variable performance, with the highest accuracy reaching 65.33%. The study contributes to the field by demonstrating the application of Hu Moments in plant classification and by assessing the K-NN algorithm's effectiveness in this context. These findings offer insights into the potential of combining image processing techniques with machine learning for accurate plant classification, paving the way for further research in the area.

show abstract

“…Where 𝑓(𝑥, 𝑦) is the pixel intensity at (𝑥, 𝑦), and (𝑥̅ , 𝑦 ̅) is the centroid of the image [15], [16].…”

Section: Image Segmentation: Cannymentioning

confidence: 99%

Automated Classification of Empon Plants: A Comparative Study Using Hu Moments and K-NN Algorithm

Hayatou Oumarou,

Rismayanti

2024

ijodas

View full text Add to dashboard Cite

show abstract

“…Think about a medical robot that can correctly classify a health question and give a possibly lifesaving answer, or a virtual tourist helper that can tell the difference between questions about food and questions about historical sites. It's not just handy that the good effects happen; they often have big effects [8], [9], [10]. However, the complexity of human language, which includes subtleties in syntax, meaning, and pragmatics, makes it very hard to get very accurate question classification [11], [12].…”

Section: Introductionmentioning

confidence: 99%

“…Support Vector Machines, Random Forests, and other machine learning models have been used for this, but new developments in deep learning and transformer models like BERT, RoBERTa, and ELECTRA have shown that they work even better than expected [13]. These models are very good at understanding the meanings and contexts of words and sentences, which is a key part of question classification [1], [14], [15], [16] and [17]. Here, we show a new method that combines three strong tools: the ELECTRA model for contextual embeddings based on transformers; Global Vectors for Word Representation (GloVe) for creating semantically rich word vectors; and Long Short-Term Memory (LSTM) networks for capturing sequence dependencies.…”

Section: Introductionmentioning

confidence: 99%

An Ensemble Approach to Question Classification: Integrating Electra Transformer, GloVe, and LSTM

Aburass,

Dorgham,

Rumman

2024

IJACSA

View full text Add to dashboard Cite

Natural Language Processing (NLP) has emerged as a critical technology for understanding and generating human language, with applications including machine translation, sentiment analysis, and, most importantly, question classification. As a subfield of NLP, question classification focuses on determining the type of information being sought, which is an important step for downstream applications such as question answering systems. This study introduces an innovative ensemble approach to question classification that combines the strengths of the Electra, GloVe, and LSTM models. After being tried thoroughly on the well-known TREC dataset, the model shows that combining these different technologies can produce better outcomes. For understanding complex language, Electra uses transformers; GloVe uses global vector representations for wordlevel meaning; and LSTM models long-term relationships through sequence learning. Our ensemble model is a strong and effective way to solve the hard problem of question classification by mixing these parts in a smart way. The ensemble method works because it got an 80% accuracy score on the test dataset when it was compared to well-known models like BERT, RoBERTa, and DistilBERT.

show abstract

Performance Evaluation of Swin Vision Transformer Model Using Gradient Accumulation Optimization Technique

Aburass,

Dorgham

2023

Lecture Notes in Networks and Systems

View full text Add to dashboard Cite

Enhancing Convolutional Neural Network using Hu’s Moments

Cited by 8 publications

References 26 publications

Automated Classification of Empon Plants: A Comparative Study Using Hu Moments and K-NN Algorithm

Automated Classification of Empon Plants: A Comparative Study Using Hu Moments and K-NN Algorithm

An Ensemble Approach to Question Classification: Integrating Electra Transformer, GloVe, and LSTM

Performance Evaluation of Swin Vision Transformer Model Using Gradient Accumulation Optimization Technique

Contact Info

Product

Resources

About