• Journal of the Korea Convergence Society
• /
• v.11 no.11
• /
• pp.33-39
• /
• 2020

In the field of Natural Language Processing, Various research such as Translation, POS Tagging, Q&A, and Sentiment Analysis are globally being carried out. Sentiment Analysis shows high classification performance for English single-domain datasets by pretrained sentence embedding models. In this thesis, the classification performance is compared by Korean E-commerce online dataset with various domain attributes and 6 Neural-Net models are built as BOW (Bag Of Word), LSTM[1], Attention, CNN[2], ELMo[3], and BERT(KoBERT)[4]. It has been confirmed that the performance of pretrained sentence embedding models are higher than word embedding models. In addition, practical Neural-Net model composition is proposed after comparing classification performance on dataset with 17 categories. Furthermore, the way of compressing sentence embedding model is mentioned as future work, considering inference time against model capacity on real-time service.

• Journal of Korea Multimedia Society
• /
• v.18 no.4
• /
• pp.483-491
• /
• 2015

Data hiding is one of protective methods that can authenticate the completeness of digital information and protect intelligent property rights and copyright through secret communications. In this paper, we propose a data hiding method using pixel-value modular operation that has a high capacity while maintaining a good visual quality. The proposed method can embed secret data on the every pixel of a cover image by modular operation. The experiment results demonstrate that the proposed method has a high capacity and good visual quality where the embedding capacity is 91,138 bytes, the PSNR is 47.94dB, and the Q index is 0.968.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.12
• /
• pp.156-163
• /
• 2012

Recently, Information Hiding Technology are becoming increasingly demanding in the field of international security, military and medical image This paper proposes data hiding technique utilizing parity checker for gray level image. many researches have been adopted LSB substitution and XOR operation in the field of steganography for the low complexity, high embedding capacity and high image quality. But, LSB substitution methods are not secure through it's naive mechanism even though it achieves high embedding capacity. Proposed method replaces LSB of each pixel with XOR(between the parity check bit of other 7 MSBs and 1 Secret bit) within one pixel. As a result, stego-image(that is, steganogram) doesn't result in high image degradation. Eavesdropper couldn't easily detect the message embedding. This approach is applying the concept of symmetric-key encryption protocol onto steganography. Furthermore, 1bit of symmetric-key is generated by the self-reference of each pixel. Proposed method provide more 25% embedding rate against existing XOR operation-based methods and show the effect of the reversal rate of LSB about 2% improvement.

• Journal of Korea Multimedia Society
• /
• v.16 no.5
• /
• pp.577-590
• /
• 2013

Shamir's (k,n)-threshold secret sharing scheme is not secure against cheating by attacker because the signature of participants is omitted. To prevent cheating, many schemes have been proposed, and a proactive secret sharing is one of those. The proactive secret sharing is a method to update shares in the secret sharing scheme at irregular intervals. In this paper, a proactive image secret sharing scheme over $GF(2^8)$ is proposed for the first time. For the past 30 years, Galois field operation is widely used in order to perform the efficient and secure bit operation in cryptography, and the proposed scheme with update phase of shadow image over $GF(2^8)$) at irregular intervals provides the lossless and non-compromising of secret image. To evaluate security and efficiency of images (i.e. cover and shadow images) distortion between the proposed scheme and the previous schemes, embedding capacity and PSNR are compared in experiments. The experimental results show that the performances of the embedding capacity and image distortion ratio of the proposed scheme are superior to the previous schemes.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2001.11a
• /
• pp.699-702
• /
• 2001

본 논문에서는 비트 플래인 상에서 지역적인 복잡도를 이용하여 커버 이미지를 분할한 후 비밀 데이터를 삽입하는 스테가노그라피 방법을 연구하였다. 이 방법은 복잡도를 이용하여 비트 플래인 이미지를 informative 영역과 noise-like 영역으로 나누고, noise-like 영역에 데이터를 삽입한다. 삽입되는 데이터가 간단하다면 image conjugation을 적용하여 복잡한 형태로 만들어 커버 이미지에 삽입한다. 본 연구에서는 삽이 용량을 증가시키기 위해 복잡도를 모든 비트 플래인에 적용시키지 않고, 선택적으로 적용하여 46%의 최대 삽입용량과 화질의 증가를 얻을 수 있었다.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.1807-1810
• /
• 2003

Steganography based on Just Noticeable Difference(JND) has been used for natural images. However, it has been recognized to have defects for the non-natural images such as scanned text images, cartoons, etc. In this paper, an alternative method is proposed to improve this problem. A new scheme is designed specially for the non-natural images. Instead of JND, Noise Visibility Function(NVF) is used. NVF value and edge strength value of each pixel ate combined to decide the embedding data capacity and the visibility of data embedded images have been improved specially for the non-natural images.

• 한국정보통신설비학회:학술대회논문집
• /
• 2006.08a
• /
• pp.274-279
• /
• 2006

대부분의 컬러 영상들은 코드북의 일부분만으로 표현할 수 있는 영상의 영역이 상당히 크므로 이러한 특징을 이용하는 효율적인 Reversible 워터마킹 기법을 제안한다. 팔레트(컬러맵) 영상에서의 워터마킹 기술에 대한 연구는 많이 있지만 Reversibility에 대한 관심은 상대적으로 낮았다. 제안하는 기법은 가상적으로 확장한 코드북에 의한 압축 영역에서 동작한다. 확장된 코드북은 높은 Payload 용량을 가진다. Reversibility를 구현하면서 왜곡(Distortion)과 삽입용량(Embedding capacity)사이의 절충점을 실험을 통해 보인다.

• 한국정보통신설비학회:학술대회논문집
• /
• 2008.08a
• /
• pp.201-204
• /
• 2008

In this paper a new steganographic method is presented with minimum distortion. This paper focused on DCT rounding error and optimized that in a very easy way, resulting stego image has less distortion than other existing methods. The proposed method compared with F5 steganography algorithm, and the proposed method achieved better performance. This paper considered the DCT rounding error for lower distortion with possibly higher embedding capacity.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.979-981
• /
• 2011

In this paper, a new scheme to improve both the capacity and distortion performance for error-prediction method is proposed. For every triangle mesh, two vertices were selected to predict the rest vertex, and the prediction error distances which were vertical and paralleled to the edge between these two vertices would be used to embed two units of secret date amount. We sort the meshes before embedding according to predicted error in order to decrease the distortion. Experiment results show that our approach increase the capacity and decrease the distortion as compared to the original algorithm.

• Journal of KIISE:Software and Applications
• /
• v.37 no.6
• /
• pp.417-429
• /
• 2010

Reversible watermarking inserts watermark into digital media in such a way that visual transparency is preserved and then enables to restore the original media from the marked one without any loss of media quality. This watermarking can be applied to quality-sensitive imaging such as medical imaging, military imaging, remote-sensing imaging, and precious artwork, where the original media should be preserved during image processing and analysis. In this paper, a reversible image watermarking technique that embeds message bits by modifying the differential histogram of adjacent pixels is presented. In order to satisfy both high embedding capacity and visual quality, the proposed technique exploits the fact that adjacent pixels in the image have highly spatial correlation. Also, we prevent overflow/underflow problem and salt-and-pepper artifacts by employing a predicted error compensation scheme. Through experiments using various test images, we prove that the presented technique provides perfect reversibility and high embedding capacity, while maintaining the induced-distortion low.