cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Widi Aribowo
Contact Email
widiaribowo@unesa.ac.id
Phone
+62811307761
Journal Mail Official
vubeta@unesa.ac.id
Editorial Address
Jl. Prof. Moch Yamin, Ketintang, Kec. Gayungan, Surabaya, Jawa Timur 60231
Location
Kota surabaya,
Jawa timur
INDONESIA
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science
Published by Universitas Negeri Surabaya
ISSN : -     EISSN : 30640768     DOI : https://doi.org/10.26740/vubeta.v1i1
Core Subject : Science, Engineering,
Computer Science & IT Engineering Mechanical Engineering Transportation
Vokasi Unesa Bulletin Of Engineering, Technology and Applied Science is a peer-reviewed, Quarterly International Journal, that publishes high-quality theoretical and experimental papers of permanent interest, that have not previously been published in a journal, in the field of engineering, technology, and applied sciences that aim to promote the theory and practice of Engineering, Technology And Applied Science.
Arjuna Subject : Teknik (semua kategori) - Teknik (Lain-Lain)
Articles 101 Documents
 7   8   9   10   11 
Zero-Shot Super-Resolution as a Test-Time Enhancer for Cross-Crop Plant Disease Recognition Malam, Sani Saminu Saleh; Ibrahim, Yusuf; Haruna, Zaharuddeen; Yusuf, Shehu Mohammed
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.45798

Abstract

Accurate plant disease diagnosis is central to precision agriculture, yet real-world performance degrades under blur, low resolution, and domain shift, weakening zero-shot recognition of unseen diseases. This paper investigates the integration of Coordinate Attention (CA) and Zero-Shot Super-Resolution (ZSSR) as test-time plug-ins to a standard Zero-Shot Learning (ZSL) pipeline without using any target labels. Using Plant Village tomato to potato transfer, each target image is super-resolved via a compact, self-supervised SR CNN (50 inner steps with self-ensemble and back-projection) and then standardized to 224×224×3 before feature extraction with MobileNetV2 (global average pooling). A lightweight CA module enhances spatial channel attention, focusing on lesion regions. The visual embeddings (1280-D) are projected into a 300-dimensional, L2-normalized semantic space through a dense, BN, ReLU to dropout head, and class logits are computed as cosine similarity to Word2Vec prototypes. On the target (potato) test set, the proposed ZSL + CA + ZSSR model achieved 86.33% accuracy, outperforming both ZSL + ZSSR (79.04%) and the ZSTL benchmark (78.34%, VGG16 + Triplet + DAC-300). Confusion matrices show fewer PEB↔PLB and PH to diseased confusions, while training curves exhibit faster, more stable convergence when ZSSR and CA are jointly applied. These results indicate that per-image, test-time ZSSR with CA attention sharpens lesion cues and enhances cross-crop transfer, providing a lightweight, label-free pathway to improved field robustness and diagnostics.
Performance evaluation of the fast forward quantum optimization algorithm in digital image clustering Singh, Sanjeev Kumar; Singh, Pawan Kumar
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.46206

Abstract

The primary objective of clustering in image analysis is to establish a meaningful correspondence between image features and clusters. This process is instrumental in extracting higher-level semantic information from digital images. In this study, we propose a novel image clustering approach that integrates the fast forward quantum optimization algorithm (FFQOA) with the K-means clustering (KMC) algorithm, forming a hybrid method referred to as FFQOA + KMC. The FFQOA + KMC initiates clustering based on the grayscale values of images using KMC and then refines the clustering outcome through FFQOA to achieve optimal segmentation. Subsequently, FFQOA + KMC is applied to several benchmark grayscale images, with results compared to those from alternative clustering techniques. Experimental findings confirm the robustness and superiority of FFQOA + KMC through both visual inspections and statistical metrics
Benz Limit Optimal Design for Double Fed Induction Generator and Kundru’s Multimachine integration Abel E. Airobman; Sabo Aliyu2; Musa, Engr.
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.46268

Abstract

The study explores the integration of an optimal Benz limit Doubly Fed Induction Generator (DFIG) with IEEE Kundur's test multi-machine power system, emphasizing the use of meta-heuristic algorithms and controllers.his work is scoped at the DFIG power coefficient, output voltage, and phase angle oscillations during integration. The controllers’ performances were compared with three techniques: the hippopotamus (HO), Sine cosine (SC), and Morth flame (MFO) algorithms to verify the competence of the proposed method in achieving better system stability. To improve the proposed Hybrid Multi-source integration of DFIG, Hydrogen Fuel Cell (HFC) to augment (Wind, Solar cell, Battery energy storage system), the proposed work presents the mathematical formulation of DAE, the designed models, and the implementation of wind aerodynamic/mechanical coupling shaft. ODE as solver in MATLAB 2021a Simulink environment as presented. The results presented an optimal Benz limit for the blade tip speed ratio  = 8.1, blade pitch angle =0, rotor power coefficient =  =0.48, and turbine output power  =5 MW described by equation 7. A symmetrical fault was set up on bus 2 at t = 1 second; the governor load reference increased by 1%, the system loading by 1%; and a nonlinear time-domain simulation was carried out on the integrated network to assess controllers’ robustness Likewise, the result validates the usefulness of the proposed SC, HO, and MFO tuned Tilt, PID for DFIG output voltage and phase angle control that outperforms the traditional MFO tuning techniques in terms of resilience, efficiency, and convergence.
A Literature Review on Brain Tumour Detection Approaches Using MRIs Ajay; Singh, Pritpal
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.46521

Abstract

Brain tumours are among the most common malignant tumours, making their accurate detection and precise evaluation crucial for effective treatment planning and strategic regimens. Recent advancements in machine learning (ML) and deep learning (DL) have significantly increased tumour identification precision, enabling the automatic pro cessing of complex imaging data and substantially reducing the needfor time-consuming manual intervention. However, persistent challenges in automated detection approaches stem from pervasive imaging artifacts, variations in image quality, and diverse tumor appearances. This comprehensive review addresses these challenges by highlighting key innovations and their clinical relevance across various automated approaches, including clustering, soft computing, and deep learning techniques for the classification and segmentation of brain tumours using magnetic resonance imaging (MRI). Furthermore, we synthesize the quantitative results of state-of-the-art models, summarizing performance measures such as the Dice Score and Sensitivity. Ultimately, this review outlines the critical future research pathways necessary to effectively address remaining obstacles and enhance the precision of automated segmentation and classification.
Development and Performance Evaluation of Silicone-Based Hydrophobic Coatings for Anti-Soiling Applications on Solar PV Glass Thaiubudeen, Ahmad A.; Sulaiman, Shaharin A.; Awais, Syed Awais Ali
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.48779

Abstract

Soiling is a major cause of performance degradation in solar photovoltaic (PV) systems, particularly in tropical environments characterized by high humidity and persistent dust accumulation. This study presents the development and experimental evaluation of silicone-based hydrophobic coatings designed to enhance the anti-soiling and water-repellent properties of PV glass substrates. Three coating formulations with varying silicon dioxide (SiO2) nanoparticle contents, 2.03 wt% (HC1), 3.34 wt% (HC2), and 0 wt% (HC3), were prepared and systematically characterized. Coating performance was assessed using water droplet mobility tests, static water contact angle (WCA) measurements, and controlled dust-accumulation experiments. Among the formulations, HC2 exhibited the best overall performance, achieving the highest average WCA (94.3°), fastest droplet runoff time (0.8 s), and lowest dust accumulation (4.4 mg). The results confirm that increasing SiO2 nanoparticle concentration enhances surface hydrophobicity and reduces dust adhesion by modifying surface roughness. These findings highlight the potential of optimized silicone-SiO2 coatings as a cost-effective, passive anti-soiling solution for improving the operational efficiency of solar PV systems, particularly in dust-prone environments.
Machine Learning Models for DDoS Attack Detection: A Systematic Literature Review Chinyere Chioma Isiekwene; Nureni Ayofe Azeez; Solomon A. Akinboro; Oladipupo Sennaike
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.40146

Abstract

The study aims to present a detailed analysis of different machine learning models used in the detection of distributed denial of service (DDoS) attacks. The report adopted the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) style to determine the research domain, established a search list, and analyzed all the selected articles from scientific databases such as IEEE, Springer, Elsevier, MDPI, SSRN-JETIR, Wiley online-library, and Google Scholar to meet eligibility criteria. A total of 6560 articles were retrieved, and 75 were deemed eligible for study. The review identified seven subject categories in the literature review, and the results show that 48% of the reviewed papers were from Elsevier (Science Direct), IEEE covered 20%, Springer covered 16%, while MDPI count was 10.67%. 2023 had the highest number of paper sources, followed closely by 2022, then 2024. The study reveals the milestone achieved in the use of machine learning models in detecting distributed denial of service attacks alongside the existing gap in the application of these models.
Estimation of R404A Refrigerant Cooling Load and System Efficiency of a Small-Scale Cold Box for Strenghten The Fisheries Cold Chain in Timbolsloko, Demak Hanityo Adi Nugroho; Samsudi R; Rubijanto JP; M. Ahnaf F; Aulia SS; Aditano YR
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.40752

Abstract

The cooling system of the fishing boat is extremely crucial in maintaining the quality of the catch during storage. In this study, the system performance and cooling load are analyzed in a box of dimensions 1.8 m × 2.4 m × 1.2 m with R404A refrigerant at an evaporation temperature of -21.8°C and a condensation temperature of 37.6°C. Total daily cooling load is 15.504 kWh, which includes heat transmission load of 5.653 kWh, product load of 6.656 kWh, internal load of 0.542 kWh, equipment load of 1.61 kWh, and infiltration load of 1.043 kWh. System performance by pressure-enthalpy (P-h) analysis provides actual Coefficient of Performance (COP) as 4 with 94% refrigeration efficiency. The 100 mm polyurethane insulation utilized are enhance energy efficiency and environmental sustainability. This cooling system design was found to be optimal for maintaining the box temperature at -20°C with high efficiency, serving to benefit the small-scale fisheries sector under tropical conditions.
Flawed Deployment of AI Sensor Technologies and Tools in National Security and Biodefense Executive Recruitment Darrell Norman Burrell; Patricia Haley
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.43923

Abstract

Artificial intelligence (AI) and sensor-enabled technologies are reshaping recruitment and human resources (HR) management by enabling automated, data-driven candidate evaluation. However, sensor-driven AI systems, such as facial analysis, voice recognition, and biometric monitoring, pose significant ethical and operational risks, particularly the perpetuation of historical biases and opaque decision-making processes. This study investigates these tensions through qualitative analysis of expert interviews with AI developers, HR professionals, and diversity, equity, and inclusion (DEI) strategists, coupled with real-world case examples, including a biodefense firm whose vision-based AI system unintentionally excluded qualified candidates. Findings reveal that while AI-sensor platforms offer efficiency and personalized experiences, they can amplify bias, obscure accountability, and challenge legal compliance if not carefully designed and governed. Participants highlighted urgent needs for algorithmic transparency, human oversight, and inclusive system design to mitigate these risks. In response, this study proposes a human-centered framework for the ethical deployment of AI-sensor technologies in hiring, emphasizing continuous bias auditing, clear governance structures, and regulatory alignment. Ultimately, it argues that the transformative potential of intelligent sensing in HR depends not only on technical sophistication but on embedding these tools within sociotechnical systems committed to fairness, accountability, and inclusion.
Biomedical colors images watermarking scheme based on LSB, Henon map and ECKBA Noura Alexendre; Fotsing Kuetche; Welba Colince; Simo Thierry; Ntsama Eloundou Pascal
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.44096

Abstract

With rapid digitalization in healthcare, biomedical images like X-rays, CT, MRI, and ultrasound scans are routinely transmitted, stored, and shared across hospital networks and telemedicine platforms. Ensuring data security, authenticity, and patient privacy during this process is a major challenge. Unauthorized access, modification, or duplication of images can result in diagnostic errors, legal issues, and loss of patient trust. Researchers have developed algorithms for image watermarking (embedding copyright or authentication information) and image encryption (scrambling data to protect it) to address these concerns. In this article, we present an algorithm that combines watermarking with encryption to enhance security and ensure confidentiality for medical images. Our approach centers on a blind hybrid watermarking technique. 'Blind' means the original image is not needed to extract the watermark, and 'hybrid' refers to combining techniques. This method specifically uses LSB (Least Significant Bit) embedding with a text image as the watermark. For encryption, chaotic sequences generated by the Henon map (a mathematical system used to generate pseudo-random numbers) power selective encryption, while the Enhanced 1D Chaotic Key Based Algorithm (ECKBA) is used for image encryption. The main advantage is our method's ability to generate a large space of encryption keys, critical for resisting brute-force attacks. Experimental and planned results demonstrate the robustness of our algorithm against common attacks and its watermark's invisibility to the human eye
Leveraging State-of-the-art Deep Learning Advancements for Emotion Detection: A Comprehensive Review and Insights Krishna Kant
Vokasi UNESA Bulletin of Engineering, Technology and Applied Science Vol. 3 No. 2 (2026): (In Progress)
Publisher : Universitas Negeri Surabaya or The State University of Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26740/vubeta.v3i2.46974

Abstract

Emotion recognition is a fundamental aspect of affective computing, focusing on identifying and interpreting human emotional states.  Among various modalities, facial emotion recognition has gained significant attention due to its non-intrusive nature and extensive applicability across domains such as e-learning, healthcare, marketing, e-commerce, and psychology. A wide range of approaches has been employed to address the challenges inherent in facial emotion classification. There remains a lack of a holistic, structured framework that critically evaluates both the advantages and shortcomings of deep networks while introducing attention-based and Transformer-driven models. Therefore, to address this gap, this paper presents a systematic review of peer-reviewed FER studies of deep learning models published between 2022 and 2025. This paper presents the study of advanced deep learning architectures for facial emotion detection, emphasizing the predominance of Deep Learning models including Transformer-based architectures, hybrid CNN–Transformer models, spatiotemporal learning approaches, and novel attention mechanisms. This research work provides analysis of deep learning model architectures, learning strategies, datasets, evaluation protocols, and performance metrics reported in state-of-the-art FER research. It identifies common issues including computational complexity, real-world robustness, generalization across datasets, and data imbalance. It also analyzes current research challenges, limitations and their practical significance. Furthermore, this research work identified and discussed the possible opportunities, unresolved issues of human facial emotion recognition and provided the future directions. The objective of this study is to provide actionable insights for researchers and practitioners, guiding future research toward more robust, accurate, and interpretable FER systems.

Page 10 of 11 | Total Record : 101

 7   8   9   10   11 

Filter by Year

2024 2026


Reset
Filter By Issues
All Issue Vol. 3 No. 2 (2026): (In Progress) Vol. 3 No. 1 (2026) Vol. 2 No. 3 (2025) Vol. 2 No. 2 (2025) Vol. 2 No. 1 (2025) Vol. 1 No. 3 (2024) Vol. 1 No. 2 (2024) Vol. 1 No. 1 (2024)