Research Cell: An International Journal of Engineering Sciences

Jaspreet Kaur ¹, Jagroop Kaur ¹

Affiliations
1 Deptt. of CE, Punjabi University, Patiala, IN

Abstract

This paper presents the hiding of text in video using haar wavelet transform in particular frame and BCH codes. Haar wavelet transform is applied to get the low frequency sub-band in an image to hide data. The frequency sub-bands are (LL, LH, HL and HH). BCH codes are used to encrypt the data. The work of various researchers is discussed about video steganography and their techniques for embedding and extraction of data. With the advancement of technology and multimedia information, Videos and digital images are increasing very quickly. Steganography is hiding private or secret data within a carrier in invisible manner. A lot of information can be stored in video files. It contains number of frames played over a period of time. Information can be stored in any or number of frames. Moreover, Attacker or Human eye cannot detect the presence of message in video because video frame is only visible for a fraction of time.

Keywords

Steganography, Cryptography, Encryption Algorithm.

Full Text

   

Manpreet Kaur ¹, Jagroop Kaur ²

Affiliations
1 Computer Science Department, UCOE, Punjabi University, Patiala, IN
2 Department of Computer Science, UCOE, Punjabi University, Patiala, IN

Abstract

In the recent years, an exponential increase of using new linguistic feature Acronyms in the available electronic information documents to produce shorter text causes a big necessity to quickly understand large volumes of data. In the recent years, Automatic Recognition of Acronym-Definition Pairs has gained a One challenge derives from the common and uncontrolled use of acronyms in the medical and political field. In this paper we have developed an Algorithm for Extracting Acronym Definitions from Gurmukhi Text by using First-Letter matching technique. For improving the recall we have provided Common Acronym-Definition Pairs database. We tested our system over two types of data- Medical and Political data. We have applied three standard measures 1) Recall 2) precision 3) F-Score. The proposed system achieved 86% recall, 86.04% precision, yielding F-score of 86.01% for Medical document and 86.53% recall, 95.55% precision, yielding F-score of 90.81% for Political document by using Database of common Acronym-Definition pairs.

Keywords

Acronyms, Acronym identification, Abbreviation, Tokenization, Normalization, Recall, Precision, Medline.

Full Text

   

Mandeep Kaur ¹, Jagroop Kaur ²

Affiliations
1 Baba Farid College of Engineering and Technology, Bathinda, Punjab, Pin Code-151001, IN
2 University College of Engineering, Punjabi University, Patiala, Punjab, Pin Code-147002, IN

Abstract

As the internet is growing rapidly, this has resulted in large amount of information. Text summarization provides shorthand version for such information, which is no longer than half of the original text. This paper proposes a system for detection and removal of Deadwood in summaries for Punjabi language. Deadwood means word or phrase that can be omitted without loss in meaning. Removing it shortens and clarifies the summary. The first step in this process is preprocessing which consists of sentence segmentation and removal of Punjabi stop words and then in the second step weight is assigned to the sentences in the source text .We used five different features for the assignment of weight to the sentences. In the next step the highest scoring sentences are selected to form the summary. In the last step the Deadwood is eliminated and removed from the summary.

Keywords

Deadwood, Phrase, Summary.

Full Text