RASEEDA HAMZAH

Dr. RASEEDA HAMZAH

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
5 June 2024 in IEEE Transactions on Components, Packaging and Manufacturing Technology
Verified
Journal Article
Closed access
ALAng Li
Ang Li
RHRASEEDA HAMZAH
RASEEDA HAMZAH
SRSiti Khatijah Nor Abdul Rahim
Siti Khatijah Nor Abdul Rahim
YGYousheng Gao
Yousheng Gao
Traditional manual detection for solder joint defect is no longer applied during industrial production due to low efficiency, inconsistent evaluation, high cost and lack of real-time data. A new approach has been proposed to address the issues of low accuracy, high false detection rates and computational cost of solder joint defect detection in surface mount technology of industrial scenarios. The proposed solution is a hybrid attention mechanism designed specifically for the solder joint defect detection algorithm to improve quality control in the manufacturing process by increasing the accuracy while reducing the computational cost. The hybrid attention mechanism comprises a proposed enhanced multi-head self-attention and coordinate attention mechanisms increase the ability of attention networks to perceive contextual information and enhances the utilization range of network features. The coordinate attention mechanism enhances the connection between different channels and reduces location information loss. The hybrid attention mechanism enhances the capability of the network to perceive long-distance position information and learn local features. The improved algorithm model has good detection ability for solder joint defect detection, with mAP reaching 91.5%, 4.3% higher than the You Only Look Once version 5 algorithm and better than other comparative algorithms. Compared to other versions, mean Average Precision, Precision, Recall, F1 score, and Frame per Seconds indicators have also improved. The improvement of detection accuracy can be achieved while meeting real-time detection requirements.

4 August 2023 in Journal of Advanced Research in Applied Sciences and Engineering Technology
Verified
Journal Article
Open access
PDF file
LALi Ang
Li Ang
RHRASEEDA HAMZAH
RASEEDA HAMZAH
SRSiti Khatijah Nor Abdul Rahim
Siti Khatijah Nor Abdul Rahim
GYGao Yousheng
Gao Yousheng
KSKhyrina Airin Fariza Abu Samah
Khyrina Airin Fariza Abu Samah
Surface Mount Technology (SMT) pin solder joint defects are hard to detect since the joints are smaller and denser and have high similarity between defect samples and standard samples in solder joint images. We propose an improved YOLOv5 defect detection algorithm embedding Cascade Shuffle Space to Depth (CSSD), Coordinate Attention (CA) mechanism module, and K-means++ algorithm. The proposed improved Yolov5 significantly impacts the loss and model parameter reduction and higher positioning precision of the defect location on the disk. The optimum anchor box produces better clustering and stability. Compared with the original YOLOv5 under the same test conditions, the method in this paper improves the precision by 12.2%, recall by 9.4%, mAP by 9.0%, and model parameters reduced by 1.3M. In conclusion, the experimental results show that the algorithm proposed in this paper has a better detection effect and a smaller parameter scale. It also can better meet the defect detection and model deployment in the actual industrial production environment.

15 April 2025 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
ALAng Li
Ang Li
RHRASEEDA HAMZAH
RASEEDA HAMZAH
YGYousheng Gao
Yousheng Gao
Underwater sonar imagery is characterized by small target sizes and low resolution, which can result in detection failures or false positives. To counteract these challenges, we introduce the Under Water Sonar Detection Transformer (US-DETR), an underwater sonar object detection model derived from the Real-Time Detection Transformer (RT-DETR) framework, incorporating attention-based feature fusion. US-DETR includes a novel Enhanced Feature Interaction module (EFI), which enhances the feature extraction network’s ability to perceive global information of the detected target. Additionally, we propose a novel Non-local Attention Feature Fusion (NAFF) module to heighten the network’s sensitivity to the spatial relationships between feature channels across different scales, thereby enhancing its channel position and global information awareness. Experiments are conducted on a benchmark underwater sonar image dataset. Experimental results show that compared to RT-DETR, US-DETR achieves a 2.2% higher mean Average Precision (mAP) and a 2.1% higher F1 Score compared to RT-DETR. The model also strikes an effective balance between detection speed and accuracy, achieving real-time performance of 126 FPS, which can meet the real-time requirements in industrial production.

1 September 2022 in IAES International Journal of Artificial Intelligence (IJ-AI)
Verified
Journal Article
Open access
PDF file
KSKhyrina Airin Fariza Abu Samah
Khyrina Airin Fariza Abu Samah
NWNurqueen Sayang Dinnie Wirakarnain
Nurqueen Sayang Dinnie Wirakarnain
RHRASEEDA HAMZAH
RASEEDA HAMZAH
NMNor Aiza Moketar
Nor Aiza Moketar
LRLala Septem Riza
Lala Septem Riza
ZOZainab Othman
Zainab Othman
This study explicitly discusses helping job seekers predict salaries and visualize job vacancies related to their future careers. Jobstreet Malaysia is an ideal platform for discovering jobs across the country. However, it is challenging to identify these jobs, which are organized according to their respective and specific courses. Therefore, the linear regression approach and visualization techniques were applied to overcome the problem. This approach can provide predicted salaries, which is useful as this enables job seekers to choose jobs more easily based on their salary expectations. The extracted Jobstreet data runs the pre-processing, develops the model, and runs on real-world data. A web-based dashboard presents the visualization of the extracted data. This helps job seekers to gain a thorough overview of their desired employment field and compare the salaries offered. The system’s reliability as tested using mean absolute error, the functionality test was performed according to the use case description, and the usability test was performed using the system usability scale. The reliability results indicate a positive correlation with the actual values. The functionality test produced a successful result, and a score of 96.58% was achieved for the system usability scale result, proving the system grade is ‘A’ and usable.

1 March 2023 in IAES International Journal of Artificial Intelligence (IJ-AI)
Verified
Journal Article
Open access
PDF file
SISHAFAF IBRAHIM
SHAFAF IBRAHIM
KSKhyrina Airin Fariza Abu Samah
Khyrina Airin Fariza Abu Samah
RHRASEEDA HAMZAH
RASEEDA HAMZAH
NANurul Amira Mohd Ali
Nurul Amira Mohd Ali
RARaihah Aminuddin
Raihah Aminuddin
Glioblastoma multiforme (GBM) is a high-grade brain tumor that is extremely dangerous and aggressive. Due to its rapid development rate, high-grade cancers require early detection and treatment, and early detection may possibly increase the chances of survival. The current practice of GBM detection is performed by a radiologist; due to the enormous number of cases, it is nevertheless tedious, intrusive, and error-prone. Thus, this study attempted a substantial adaptive artificial bee colony (a-ABC) algorithm implementation in providing a non-invasive approach for GBM detection. The basic statistical intensity-based analysis of minimum (minGL), maximum (maxGL), and mean (meanGL) of grey level data was employed to investigate the GBM's feature properties. The a-ABC's performance for adaptive GBM detection identification was evaluated using T1-weighted (T1), T2-weighted (T2), fluid attenuated inversion recovery (FLAIR), and T1-contrast (T1C) which are four different magnetic resonance imaging (MRI) imaging sequences. Hundred and twenty MRI of GBM images were assessed in total, with 30 images per imaging sequence. The overall mean of GBM detection accuracy percentage was 93.67%, implying that the proposed a-ABC algorithm is capable of detecting GBM brain tumors. Other feature extraction strategies, on the other hand, may be added in the future to enhancee the performance of feature extraction.

9 January 2024 in Journal of Advanced Research in Applied Sciences and Engineering Technology
Verified
Journal Article
Open access
PDF file
KSKhyrina Airin Fariza Abu Samah
Khyrina Airin Fariza Abu Samah
NJNur Shahirah Jailani
Nur Shahirah Jailani
RHRASEEDA HAMZAH
RASEEDA HAMZAH
RARaihah Aminuddin
Raihah Aminuddin
NANor Afirdaus Zainal Abidin
Nor Afirdaus Zainal Abidin
LRLala Septem Riza
Lala Septem Riza
Twitter has become a popular platform for the citizens of Malaysia. Twitter’s ease of expressing opinions could be used to evaluate and review Malaysian Online Food Delivery (OFD) providers. Due to competition from other OFDs in Malaysia, companies need to know customer feedback. OFD reviews are unstructured and massive, making comparisons difficult. Next, some websites evaluate OFD yet only consider pricing, delivery time, and customer experience. Customers cannot visualize the comparison based on users’ preferences, bilingual reviews, and less time-consuming to visualize OFD using the website. Thus, this study aims to design a web application system that uses Naïve Bayes to categorize Twitter sentiment analysis (SA) on Malaysia’s best OFD. It is based on customer satisfaction, visualizing the results, developing the system, and evaluating its accuracy, functioning and usability. Users can read about specific OFD by viewing Twitter SA visualization or comparing them directly. Five aspect-based SA types were presented: affordable price, promotion and discount, review rating, delivery time and condition of food delivered. Functionality testing demonstrated the accomplishment of all objectives. The training and testing data could predict OFD’s Twitter sentiment with 71.67% and 76.29% accuracy for English and Bahasa Melayu, respectively. The system’s usability produced a 94.64% average score using System Usability Scale and was considered “excellent”. Thus, it can be concluded that this study can solve the mentioned issues of OFD and ease the aspect-based comparison.