An Efficient Mathematical Method for Recovery of Tampered Image Content DOI
Tong Liu, Xiaochen Yuan, Chan–Tong Lam

et al.

2022 IEEE 10th International Conference on Information, Communication and Networks (ICICN), Journal Year: 2023, Volume and Issue: unknown, P. 782 - 786

Published: Aug. 17, 2023

To protect the integrity and reliability of images, we propose a watermarking-based approach for identifying whether image has been tampered with during transmission recovering content to its original form. The pixel values are treated as unknowns in our proposed method calculated by solving system equations using Gauss-Jordan elimination. recovery data is obtained encoding MSB matrix secret key then embedded part watermark self-recovery. remainder authentication information localizing forged region, which produced each block. Compared existing works, experimental results show superior capability.

Language: Английский

Observation of Imbalance Tracer Study Data for Graduates Employability Prediction in Indonesia DOI Open Access
Ferian Fauzi Abdulloh, Majid Rahardi, Afrig Aminuddin

et al.

International Journal of Advanced Computer Science and Applications, Journal Year: 2022, Volume and Issue: 13(8)

Published: Jan. 1, 2022

Tracer Study is a mandatory aspect of accreditation assessment in Indonesia. The Indonesian Ministry Education requires all Indonesia Universities to anually report graduate tracer study reports the government. also needed by University evaluating success learning that has been applied curriculum. One things need be evaluated level absorption graduates into working industry, so machine model assist Officials and understanding character its graduates, it can help determine curriculum policies. In this research, researcher focuses on making reliable with dataset format determined Government was obtained from Amikom University. study, SVM will tested several variants algorithm handle imbalanced data. compared SMOTE, SMOTE-ENN, SMOTE-Tomek combined detect employability graduates. test carried out K-Fold Cross Validation, highest accuracy precision results produced SMOTE-ENN value 0.96 0.89.

Language: Английский

Citations

11
An edge inspired image watermarking approach using compass edge detector and LSB in cybersecurity DOI
Zaid Bin Faheem, Danial Hanif, Farrukh Arslan

et al.

Computers & Electrical Engineering, Journal Year: 2023, Volume and Issue: 111, P. 108979 - 108979

Published: Oct. 12, 2023

Language: Английский

Citations

6
Predictive Models Using Supervised Neural Network for Pollutant Removal Efficiency in Petrochemical Wastewater Treatment DOI
Varun Geetha Mohan, Al-Fahim Mubarak Ali, Mohamed Ariff Ameedeen

et al.

2022 5th International Conference on Information and Communications Technology (ICOIACT), Journal Year: 2022, Volume and Issue: unknown, P. 116 - 121

Published: Aug. 24, 2022

The important process in wastewater treatment is the removal of pollutants, and dataset having so many features may cause difficulty training data predicting key variables. This work aims to propose set parameters through normalization techniques, feature selection AI techniques. datasets have 36 a parameter, experimental contain 628. Constant factor, Z-score, Min-max are techniques used normalize petrochemical dataset. SelectKBest, ExtraTreeClassifier, PCA, RFE for mining. Then finally done with implementation help supervised neural network technique called backpropagation (BPNN).

Language: Английский

Citations

9
Sentiment Analysis of Review Sestyc Using Support Vector Machine, Naive Bayes, and Logistic Regression Algorithm DOI

Barka Satya,

Muhammad Hasan S J,

Majid Rahardi

et al.

2022 5th International Conference on Information and Communications Technology (ICOIACT), Journal Year: 2022, Volume and Issue: unknown, P. 188 - 193

Published: Aug. 24, 2022

The growth of internet users in Indonesia experiences a very high increase every year, with the also resulting many people using social media. Sestyc is media application created by group millennial children Indonesia. This study was conducted to analyze sentiment text data form review obtained from Google Play Store. purpose this research towards sestyc and find best algorithm for classifying sentiment. used analyzing consists Support Vector Machine, Logistic regression, Naive Bayes. results class labeling on 8000 reviews total 4719 positive 3281 negative reviews. indicate that Machine has highest accuracy value compared other algorithms, where gets an value. 87.81%.

Language: Английский

Citations

9
An unsolvable pixel reduced authentication method for color images with grayscale invariance DOI Creative Commons
Xiaoyu Zhou, Wien Hong, Guangsong Yang

et al.

Journal of King Saud University - Computer and Information Sciences, Journal Year: 2023, Volume and Issue: 35(9), P. 101726 - 101726

Published: Aug. 24, 2023

This paper introduces an advanced grayscale-invariant authentication method for color images. Prior work directly utilizes the most significant bits (MSBs) of red and blue channels code generation embeds these codes into least (LSBs). Concurrently, green channel is adjusted to sustain consistent grayscale values. However, this direct utilization MSBs often leads a decline in image quality results number unsolvable pixels. To address challenges, we present new approach specifically designed combat deterioration prevalence In our method, propose MSB alteration (MSBA) technique, which modifies generate with minimal distortion. technique effectively reduces distortion Furthermore, introduce pixel substitution strategy (PSS) constraint-reduced (CRS) enhance embedding efficiency. Experimental demonstrate that outperforms prior by converting more pixels solvable ones achieving better detection results. Moreover, enhances approximately 1.5 dB, while maintaining same capacity.

Language: Английский

Citations

4
AuSR3: A new block mapping technique for image authentication and self-recovery to avoid the tamper coincidence problem DOI Creative Commons
Afrig Aminuddin, Ferda Ernawan

Journal of King Saud University - Computer and Information Sciences, Journal Year: 2023, Volume and Issue: 35(9), P. 101755 - 101755

Published: Sept. 15, 2023

This paper proposes a new block mapping technique for image authentication and self-recovery designed to avoid the tamper coincidence problem called AuSR3. The can arise when modifications an affect original its recovery data, resulting in inability recover tampered region of image. ensures that data is embedded into most distant location possible, minimizing problem. In addition, improved LSB shifting algorithm employed embed watermark consisting data. experimental result shows AuSR3 produce high-quality watermarked images across various datasets with average PSNR values 46.2 dB, which by 2.1 dB compared replacement technique. avoids up 25% tampering rates. It contributes recovered SSIM value 39.10 0.9944, respectively, on 10% rate USC-SIPI dataset.

Language: Английский

Citations

4
AuCFSR: Authentication and Color Face Self-Recovery Using Novel 2D Hyperchaotic System and Deep Learning Models DOI Creative Commons
Achraf Daoui, Mohamed Yamni, Torki Altameem

et al.

Sensors, Journal Year: 2023, Volume and Issue: 23(21), P. 8957 - 8957

Published: Nov. 3, 2023

Color face images are often transmitted over public channels, where they vulnerable to tampering attacks. To address this problem, the present paper introduces a novel scheme called Authentication and Face Self-Recovery (AuCFSR) for ensuring authenticity of color recovering tampered areas in these images. AuCFSR uses new two-dimensional hyperchaotic system modular sine-cosine map (2D MSCM) embed authentication recovery data into least significant bits image pixels. This produces high-quality output with high security level. When is detected, executes two deep learning models: CodeFormer model enhance visual quality recovered DeOldify improve colorization image. Experimental results demonstrate that outperforms recent similar schemes tamper detection accuracy, level,

Language: Английский

Citations

4
Selection of Prospective Workers Using Profile Matching Algorithm on Crowdsourcing Platform DOI
Ahmad Cucus,

Luhur Bayu Aji,

Al-Fahim Bin Mubarak Ali

et al.

2022 5th International Conference on Information and Communications Technology (ICOIACT), Journal Year: 2022, Volume and Issue: unknown, P. 122 - 126

Published: Aug. 24, 2022

The use of a crowdsourcing platform is an option to get workers who will help complete the work. Crowdsourcing process gathering work, information, or opinions from large number individuals using internet, social media, smartphone apps. Whether used for programming, design, content creation, any other task, requesters are putting their trust in unfamiliar with knowledge and have unknown histories skills. Requesters do not time resources screen all crowd's qualities, unlike employing full-time personnel. In this study, we try minimize risks faced by when namely increasing match between profile jobs offered on platform. researcher implemented matching method dataset consisting several fields that became criteria finding match. find work consist two parts, core factors secondary factors. Core Factor Criteria as skill, designation, location, factor years experience. These become variables algorithm best profiles offered. This able select worker 10,000 datasets, up 1148 people most suitable tasks And results obtained indicate increase needs requester.

Language: Английский

Citations

7
Novel Integer Hahn Transform for Real-time Lossless Image Authentication in IoT: A 32-bit Microcontroller Implementation DOI Creative Commons
Achraf Daoui, Mohamed Yamni, Paweł Pławiak

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 131044 - 131064

Published: Jan. 1, 2024

Language: Английский

Citations

1
A 3D Cuboid Image Encryption Algorithm Based on Controlled Alternate Quantum Walk of Message Coding DOI Creative Commons
Pai Liu, Shihua Zhou,

Wei Qi Yan

et al.

Mathematics, Journal Year: 2022, Volume and Issue: 10(23), P. 4441 - 4441

Published: Nov. 24, 2022

In order to solve various security risks faced by image privacy protection, we propose a 3D cuboid encryption scheme based on message-encoded controlled alternate quantum walks. Firstly, calculated the initial parameters of system and performed one-dimensional walk generate probability distribution sequence. Secondly, encoded sequence into quaternary message using multiple sets messages control model, generating amplitude matrix obtain hash through matrix. Then, was divided blocks integrated cuboid. The scrambled between layers value sequence, used complete scrambling cross-section layers. Finally, converted each pixel binary cube it perform rotation operation magnitude matrix, then XOR obtained tilted final encrypted image. simulation results show that can resist typical attacks has good performance.

Language: Английский

Citations

5