cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
-
Contact Email
-
Phone
-
Journal Mail Official
-
Editorial Address
-
Location
Kota malang,
Jawa timur
INDONESIA
PROCEEDING IC-ITECHS 2014
Published by Sekolah Tinggi Informatika dan Komputer Indonesia Malang
ISSN : -     EISSN : -     DOI : -
Core Subject : Science, Education,
Computer Science & IT Education
Arjuna Subject : -
Articles 235 Documents
 12   13   14   15   16 
Advancements in Text-to-Image Diffusion Models for Personalized Image Generation: A Review of ID-Preserving Techniques of InstantID Kurniawan, Ardhiansyah
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1566

Abstract

The evolution of text-to-image diffusion models, such as GLIDE, DALL-E 2, and Stable Diffusion, has significantly enhanced image generation capabilities. However, achieving image personalization with precise facial detail retention, minimal reference images, and reduced computational costs remains challenging. Traditional methods like DreamBooth and Textual Inversion rely on extensive fine-tuning, while techniques like IP-Adapter, which avoid fine-tuning, often compromise accuracy. Addressing these gaps, InstantID introduces a novel plug-and-play module that uses a single reference image to enable efficient identity preservation with high fidelity and flexibility. InstantID departs from conventional approaches by employing ID Embedding and an Image Adapter to enhance semantic richness and facial detail fidelity. Unlike models relying on CLIP-based visual prompts, InstantID integrates ID Embedding with ControlNet to refine the cross-attention process. This involves using simplified facial keypoints for conditional input and replacing text prompts with ID Embedding. Trained on a large-scale dataset comprising LAION-Face and additional high-quality annotated images, InstantID demonstrates superior ID retention and facial detail restoration. Notably, its performance improves with multiple reference images but remains highly effective with just one. The results highlight the effectiveness of InstantID's modular components, such as IdentityNet and the Image Adapter, in ensuring exceptional generation quality and detail retention. Although currently optimized for SDXL checkpoints, InstantID offers a scalable and efficient solution for personalized image generation. By integrating with tools like ComfyUI, it provides a seamless and accessible approach to image personalization with strong ID control and adaptability.
Utilization of YouTube as a Learning Media Algorithms of Programming for Students Informatics Engineering UNIPMA Nita, Sekreningsih; Wicaksono, Dayu Agung
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1570

Abstract

YouTube as an information sharing site is an application that is provided free of charge and is easy to access and use in various activities, on of which is for creating learning media. Learning with YouTube can improve skills and strengthen learning in the form of videos, making it easier to study and develop discussion material for Professional Educators. The purpose of using learning media via YouTube is to help Informatics students understand teaching material more easily and can motivate them to learn. The method in this research uses descriptive qualitative and collects primary and secondary data, namely students who take the 1st semester Algorithms of Programming course, class 1A, 1B, 1C totaling + 50 people and other internet users who watch videos on YouTube and library journals that are relevant to the research topic . The process and steps for making YouTube as a learning media include, 1) Making storyboards for material concepts 2) Making ppt files as discussion material 3) Making learning videos 4) Editing videos 3) Uploading videos to YouTube. The final result was +250 views, comments and likes. Whereas, learning videos can be viewed via the YouTube channel at the link https : // www.youtube.com / watch?v = bRsSFCc8kV0. The conclusion of this research is that using YouTube as a learning media makes learning more interesting, effective and very useful in the learning process and increases student motivation in learning more optimally because it can be played back to understand the material more deeply
Intelligent Waste Segregation System Using Convolutional Neural Networks for Deep Learning Applications Rahmawati, Siti Solehah Yunita; Maharani, Desy Khalida; Munada, Wina
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1579

Abstract

Efficient waste management is essential for environmental sustainability and reducing landfill burdens. This study proposes an Intelligent Waste Segregation System leveraging Convolutional Neural Networks (CNNs), specifically the VGG-16 model, to automate the classification of waste into recyclable and non-recyclable categories. The purpose of this research is to enhance waste sorting accuracy and efficiency using advanced deep learning techniques. The system employs VGG-16, pre-trained on a large dataset, and fine-tuned with a waste image dataset, enabling high precision in recognizing waste types. The methodology includes dataset preprocessing, model training, and performance evaluation using metrics such as accuracy, precision, and recall. Experimental results demonstrate that the proposed system achieves a classification accuracy of 96%, surpassing existing traditional methods. The implications of this research include improving recycling processes and reducing environmental pollution through accurate waste segregation. This system has practical applications in urban waste management and recycling facilities, providing a scalable solution to global waste challenges. The findings highlight the potential of CNN-based models, particularly VGG-16, in addressing critical environmental issues. In conclusion, the proposed system offers an effective approach to automated waste segregation, paving the way for sustainable waste management practices through deep learning applications.
Visual Identity of Tlogo Land Tourism to Enhance Brand Awareness Fernanda, Rico Alan; Wardani, Lely Surya
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1580

Abstract

Tlogo Land is a tourist destination in Malang Regency that faces issues with visual identity inconsistency, including the use of different logos across various social media platforms. This study aims to design a consistent and integrated visual identity to enhance brand awareness and attract tourists. The research employs a non-participatory approach, with data collected through direct observation of Tlogo Land and interviews with the management. The process of designing the visual identity follows the Design Thinking method, consisting of five stages: Empathize, Define, Ideate, Prototype, and Test. The results show that the designed visual identity, including the logo, colors, and other graphic elements, effectively reflects the character and values of Tlogo Land as a nature-based tourist destination. The implementation of the visual identity on media such as business cards, staff uniforms, and stamps proves to be effective in increasing recognition and creating a positive image of Tlogo Land among visitors. This study concludes that developing a consistent visual identity can be an effective strategy for brand awareness of tourist destinations, while also strengthening Tlogo Land's image as a professional and attractive tourist destination
Analysis of the Level of Satisfaction of E-Learning Users at PGRI Madiun University Using the Pieces Method. Nugrahanti, Fatim; Novietasari, Eka Resty; Fatihah, Kahfi Amani
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1581

Abstract

The development of technology is a way to quickly disseminate information to users and can help solve problems faced by an institution. The level of user satisfaction with the Universitas PGRI Madiun e-learning system uses the PIECES method (Performance, Information, Economy, Control, Efficiency, and Service). This method was chosen because it is able to identify various aspects of system performance that affect user experience. This study was conducted by collecting data through questionnaires distributed to system users, namely students and lecturers using the Googleform platform. The results of the analysis showed that the Performance and Service aspects obtained the highest satisfaction scores, which indicated that the system was running stably and providing good service to users. However, the Control and Efficiency aspects require improvement, especially regarding ease of access and management of security features. These findings are expected to be input for e-learning system managers to improve the quality of service and overall user experienceThe development of technology is a way to quickly disseminate information to users and can help solve problems faced by an institution. The level of user satisfaction with the Universitas PGRI Madiun e-learning system uses the PIECES method (Performance, Information, Economy, Control, Efficiency, and Service). This method was chosen because it is able to identify various aspects of system performance that affect user experience. This study was conducted by collecting data through questionnaires distributed to system users, namely students and lecturers using the Googleform platform. The results of the analysis showed that the Performance and Service aspects obtained the highest satisfaction scores, which indicated that the system was running stably and providing good service to users. However, the Control and Efficiency aspects require improvement, especially regarding ease of access and management of security features. These findings are expected to be input for e-learning system managers to improve the quality of service and overall user experience. Keywords E-Learning, User Satisfaction, PGRI Madiun University, Pieces Method.
Implementation Of The Iot Device For Sit-Up Counting Based On ESP 8266 Rifai, Jefri Andri; Styawati, Styawati; Wijaya, Rio
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1585

Abstract

Sit ups are a type of strength exercise that functions to strengthen muscles. Sit up movements are usually calculated manually, but as technology develops, it is necessary to innovate to build a digital technology-based sit up movement counting tool. The goal is to provide a sit up instrument that is able to accurately calculate the number of sit ups that have been done. The tool is made based on digital technology using the internet of things (IoT). In making this tool, the author uses 2 LDR sensor modules and 2 KY008 laser sensors installed to detect the shoulder. If one of the light sensors in the form of an LDR module is not disconnected from the light displayed by the laser sensor, then the series of sit up counting tools with this application will not calculate the sit up acquisition. The sit up counter tool will calculate if the sit up position is perfect and cut off the laser light, it will send the results of the sit up to the MySQL database which will later be displayed in the web server. This tool is easy to use and portable or easy to move. The way this tool works is quite easy, the user only needs to choose the type of exercise needed, then the tool will calculate the movements done and the user just needs to lie on the side of the tool and prepare to do the display according to the activity to be done. Then the tool will calculate automatically and will enter the website data.
The Effectiveness of Snort in DDoS Attack Scenarios Izzatillah, Tartila Izzatillah; Fania, Fania; Zulfata, Zulfata
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1588

Abstract

Secure communication in the digital era has become important, while on the other side, the attackers are consistently moving into more sophisticated ways to acquisition their target. DDoS was one of the common techniques used with the aim to overwhelm and make their target fail to function. To detect and limit this threat, many organizations have begun to employ various solutions. However, it was discovered that several solutions are ineffective while others require high cost for the implementation, and this has become a challenge for medium to low-sized organizations to meet their business strategy. For that, in this study we aim to introduce Snort as an open-source solution for the network detection and prevention system to determine how it undertakes and performs analysis especially in terms of accuracy and speed in relation to DDoS attacks. This study will involve a few steps: first, to establish the simulation environment; secondly, to perform the simulation with a DDoS attack to allow Snort to capture the traffic and respond according to the pre-established rules; and finally, to measure and evaluate the result. Snort as an open-source product, in nature allows the public to contribute, and that becomes the advantage compared with other commercial products especially in detecting anomalies. This will help the administrators to react more quickly by having an accurate information with an earlier warning system. Additionally, Snort is affordable, making it good choice for the organization.
Jelajah Sultra: Inovation Of UI/UX Design for Smart Tourism App with AI-Based Travel Assistant to Support SDG 8 Rahmat Saputra, La Ode Alvin; Saputra, Rizal Adi; Ramadhan, Muh. Akbar Perdana
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1589

Abstract

The Jelajah Sultra application is a conceptual digital innovation designed to promote sustainable tourism in Southeast Sulawesi by incorporating an AI-powered smart travel assistant and optimal UI/UX design principles. This initiative addresses the growing demand for user-friendly and efficient digital platforms in the tourism sector, particularly in response to increasing tourist numbers in Indonesia. The primary aim of this study is to develop a UI/UX design concept for a tourism application that supports sustainable tourism practices, enhances local economic growth, and empowers communities by improving market accessibility. Utilizing the Design Thinking method, this research involves the stages of empathizing, defining user needs, ideating, prototyping, and testing to create a user-centric design for Jelajah Sultra. The design incorporates features such as AI-powered chatbot assistance, interactive maps, and streamlined navigation to ensure a superior user experience. The usability testing resulted in a high usability score of 94.0, indicating the potential of Jelajah Sultra to facilitate seamless information delivery and effective interaction for users. This research contributes to SDG 8 by fostering local economic growth and supporting responsible tourism practices.
AI-Based Phishing Attack Detection And Prevention Using Natural Language Processing (NLP) Sospeter, Birir Kipchirchir; Odoyo, Wilfred
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1590

Abstract

Phishing attacks remain one of the most prevalent and damaging cybersecurity threats, targeting users across various communication channels such as email, social media, and SMS. Traditional phishing detection systems are often limited to email and rely on static rule-based filtering or keyword matching, making them ineffective against evolving phishing tactics. This project proposes an innovative solution that utilizes Artificial Intelligence (AI) and Natural Language Processing (NLP) to create a real-time phishing attack detection and prevention system. By analyzing the contextual language of messages across multiple platforms, the system can detect and block phishing attempts with high accuracy. The system extracts important linguistic features such as urgency, emotional tone, and anomalous patterns within text, and applies machine learning algorithms—such as Random Forest, Support Vector Machines (SVM), and deep learning models like Long Short-Term Memory Networks (LSTM)—for classification. Additionally, a feedback loop is integrated to allow the system to adapt and improve over time through active learning, ensuring the detection system evolves alongside emerging phishing techniques. This AI-based solution extends beyond traditional email phishing detection by incorporating multiple channels, including SMS and social media platforms, making it a versatile tool for individuals and businesses. The system offers automated prevention actions, such as flagging suspicious messages and alerting users, thus providing a robust defense against phishing attacks in real-time. The project's implementation aims to fill the market gap in comprehensive, multi-channel phishing detection and contribute to the growing demand for intelligent and adaptive cybersecurity solutions.
The Quantum-Assisted Fingerprint Biometrics: a Novel Approach To Fast And Accurate Feature Extraction And Synthetic Generation Sospeter, Birir Kipchirchir; Odoyo, Wilfred; Akinyi, Laura
IC-ITECHS Vol 5 No 1 (2024): IC-ITECHS
Publisher : LPPM STIKI Malang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32664/ic-itechs.v5i1.1591

Abstract

This research explores the integration of artificial intelligence (AI) and quantum computing to enhance fingerprint biometrics through improved feature extraction and synthetic fingerprint generation. Traditional fingerprint biometrics face challenges related to processing speed and scalability, particularly when managing large datasets or creating synthetic fingerprints for testing and training purposes. We propose a dual approach: using convolutional neural networks (CNNs) to extract distinctive fingerprint features—such as loops, whorls, and minutiae points—and employing generative adversarial networks (GANs) for the synthesis of high-quality fingerprint images that preserve realistic patterns and variations. To address computational limitations in processing these data-intensive tasks, we explore the use of quantum computing algorithms. Specifically, we implement a hybrid quantum-classical model, using quantum support vector machines (QSVM) for feature classification and quantum-enhanced GANs (QGAN) to speed up synthetic fingerprint generation. Preliminary results indicate that quantum-assisted models demonstrate promising efficiency gains in both feature extraction and image synthesis, potentially enabling faster processing and improved scalability compared to classical models alone.This study contributes to biometric security by providing a framework for faster, more accurate fingerprint biometrics using cutting-edge AI and quantum methodologies. The findings hold potential applications in security systems, law enforcement, and digital identity management, where real-time analysis and synthetic data generation can strengthen verification and identification processes. Future work includes optimizing quantum components for larger datasets and further refining AI models to improve the realism of generated fingerprint images.

Page 14 of 24 | Total Record : 235

 12   13   14   15   16 

Filter by Year

2014 2024


Reset
Filter By Issues
All Issue Vol 5 No 1 (2024): IC-ITECHS Vol 3 No 1 (2022): IC-ITECHS Vol 2 No 1 (2020): IC-ITECHS Vol 1 (2014): Prosiding IC-ITECHS 2014 PROCEEDING IC-ITECHS 2014