Reversible data hiding hides in an image such that the original is recoverable. This paper presents a novel embedding framework with reduced distortion called skewed histogram shifting using pair of extreme predictions. Unlike traditional prediction error schemes, where only one good used to generate histogram, proposed scheme uses predictions two histograms. By exploiting structure pixels from peak and short tail are for embedding, which decreases lesser number being shifted. Detailed...

Traditionally, body image literature has used race as a variable to explain ethnic-specific differences in satisfaction and the prevalence of eating disorders. Instead employing an explanatory variable, present study utilized qualitative method explore relationships among race, ethnicity, culture, discrimination, for African American Black women. The purpose was gain deeper understanding how gender interface with inform image. Women were recruited through community centers major metropolitan...

Recently, reversible data hiding (RDH) techniques for JPEG images have become more extensively used to combine image and authentication information conveniently into one file. Although embedding in degrades visual quality increases file size, it is proven be useful communication authentication. In this paper, a method using new coefficient selection technique proposed. The proposed scheme embeds the histogram shifting (HS) method. According number of zero AC coefficients, block ordering...

Reversible data hiding with automatic contrast enhancement methods provide an interoperable way to reduce the storage requirement for image applications: original can be recovered from enhanced without any additional information. Unlike previous work, where goal was maximize contrast, proposed method increases appropriate level using idea called brightness preservation. This is achieved by adaptive bin selection process based on brightness. Extensive experimental results verify that images...

Automatic image enhancement is increasingly becoming a popular tool for the smartphone environment. The automatically enhances right after it has been stored to enhance user's experience. But, because enhancements are subjective and dependent on image, original backed up provide recovery option. This requirement inevitably increases storage requirement. In order reduce it, novel automatic contrast based reversible data hiding (ACERDH) proposed. proposed method mimics equalization effect...

Review-based control charts that integrate sentiment analysis with traditional are emerging as effective tools for analyzing customer complaints. However, existing approaches face significant challenges due to their lack of alignment the natural characteristics online reviews, such limited review volumes or imbalanced mentions attributes. To address these challenges, which commonly encountered in real-world applications, this study proposes a novel framework review-based complaint combines...

Bitcoin exchanges rely heavily on traditional intrusion detection system to secure their system. However, this reliance has proven be high risk, since and other blockchain-based transactions are not easily reversible. Many of the attacks have shown that is enough safeguard against all possible attacks, most importantly, in some cases, it takes a long time assess damage. In paper, we first describe three types models propose mitigation using blockchain analysis for each. The proposed exploit...

Reversible data hiding in JPEG images has become an important topic due to the prevalence and overwhelming support of image format these days. Much existing work focuses on embedding using AC (quantized alternating current coefficients) maximize capacity while minimizing distortion file size increase. Traditionally, DC direct are not used for embedding, assumption that DCs cause more than ACs. However, analytic which extracts fine details as a feature, ACs is acceptable, because they...

Overtourism has given rise to conflict among various stakeholders. Accordingly, control overtourism, the public sector started implement policies. Recently, Udo off Jeju Island in South Korea begun experiencing overtourism; prevent situation from deteriorating, implemented a vehicle restriction policy. This study used cost-benefit analysis framework assess social costs and benefits of policy overtourism Udo. Through interviews relevant data documents, this classified items related that could...

Recently, deep learning techniques have been developed for source code summarization. Most existing studies simply adopted natural language processing techniques, because summarization can be considered as machine translation tasks from into descriptions. However, and its description are very different, not only in the languages of writing, but also purpose writing. There is a large semantic gap between codes programming their descriptions languages. To respond to gap, we propose two-phase...

We present a novel format preserving JPEG encryption scheme, which also losslessly reduces the files size by compressing quantized DC coefficients. The main idea behind this paper is to introduce additional lossless compression as part of reduce redundancy well encryption. result encrypted file up 3.2%.

Storytelling is the act of passing on what you want to tell other people as so interesting and true-to-life story. As study in text mining progresses express words, sentences paragraphs vector, it possible classify generate using vectors. However, has not much progressed a correct flow context grammar mining. In this paper, we propose first neural network model that learn one sentence vector mapped for representing vectors sentences. And second contents stories generating Finally evaluate...

컴퓨터 사용자가 사람인지 아닌지를 판별하는 CAPTCHA는 많은 포털 사이트에서 자동 프로그램에 의한 비정상적인 회원가입 또는 다중 로그인 방지 등을 위해 사용되고 있다. 웹 사이트들은 숫자 혹은 영어로 구성된 문자열 기반 캡챠를 대부분 사용하는데, 최근에는 OCR 기술의 발달로 단순한 텍스트 캡챠는 쉽게 무력화 된다. 이에 대한 대안으로 글자 판독을 어렵게 하기 잡음을 첨가하거나 글자를 왜곡시키는 등 다양한 시도를 하고 본 논문에서 대상으로 하는 국내 한 사이트 역시 공격자들에 의해 공격을 당하였고, 끊임없이 발전시키고 논문에서는 해당 현재 있는 자연 배경을 지닌 캡챠에 대해 분석하고, SVM을 이용한 특징 분리 후 CNN을 인식을 통해 캡챠의 취약성을 검증하였다. 실험 결과, 총 1000개의 캡챠 이미지 중 368개에 정확히 맞추었고, 이를 사용하고 새로운 버전의 안전하지 않음을 입증하였다. CAPTCHA(Completely Automated Public Turing...

Privacy and intellectual rights management is becoming an important topic. The existing social network services (SNS) have many limitations in meeting consumers' expectations of their privacy management. Metadata SNS heterogeneous susceptible to personal identification copyright information misuses. There are cases where consumers not informed how metadata shared used, while the copyrights images removed without owner's consent. In this paper, we propose two protection models called...

This study aims to analyze global research trends in the life cycle assessment (LCA) of service products. Using Scopus database, we searched for LCA published from 1970 2024 based on keywords, titles, and abstracts. A total 23,970 records were obtained, which 90.67% 2010 2024, showing a significant increase across various fields during this period, with CAGR 13.6%. However, products was found account only 0.51% (122 records) studies. Given growing importance industry its environmental...

Reversible watermarking can embed data into the cover image and extract from stego image, where original be recovered perfectly after extraction of data. Difference expansion (DE) prediction error (PEE) are two popular reversible methods. DE has advantage small distortion while PEE large embedding capacity smaller compared with pixel difference. In this paper, we proposed a novel method that combines advantages PEE, difference calculated between pixels is combined edge information near pair....

This paper presents a novel weighted sparse representation classification for face recognition with learned distance metric (WSRC-LDM) which learns Mahalanobis to calculate the weight and code testing face. The is by using information-theoretic learning (ITML) helps define better used in WSRC. In meantime, takes advantage of rule SRC proposed method classify more accurately. Extensive experiments verify effectiveness method.

Automatic contrast enhancement using reversible data hiding (ACERDH) has been found to be useful in automatic image field. Instead of just providing ability, ACERDH also equalizes the pixel histogram as a part process. This allows original recovery directly from enhanced image, without any additional data, which provides file saving feature by not having save image. In this paper, we propose two improvements and novel distortion minimization methods for [1] method. The experimental results...

Building a rentable content delivery network (CDN), p2p based CDN that supports control and is composed of hardware rented from consumers, requires significant trust between the renters consumers. The proposed method solves this using Blockchain to provide transparency in running network, proof provenance, verifiable origin used determine who can get paid for their services discourage malicious activities.