• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.3
• /
• pp.310-315
• /
• 2004

Image taken in the night can be low-contrast images because of poor environment and image transmission. We propose an algorithm that improves the acquired low-contrast image. MPEG-2 separates chrominance and illuminance, and compresses respectively because human vision is more sensitive to luminance. We extracted illumination and used K-means algorithm to find a proper crossover point automatically. We used K-means algorithm in the viewpoint that the problem of crossover point selection can be considered as the two-category classification problem. We divided an image into two subimages using the crossover point, and applied the histogram equalization method respectively. We used the index of fuzziness to evaluate the degree of improvement. We compare the results of the proposed method with those of other methods.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.25 no.6
• /
• pp.1069-1078
• /
• 2001

Pattern grading is a technique used to increase or decrease the size of a garment pattern according to the measurements in a given size chart. The original pattern is graded and laid out for cutting before mass production. This study tried to classify body types for pattern grading of jacket by applying a concept of "drop"defined as the difference between chest girth and waist girth and the difference between hip girth and waist girth for pants. Data were collected through the stratified sampling method. 138 subjects were selected out of 1,290 subjects of our sample population. Findings were as follows : 1) For pattern grading of jacket, the cell with the chest girth of 96cm and the waist girth of 87cm had the highest frequency rate and body type was 87H type and the coverage of this type was 9.52%. Then, the size specification 87-96 was the center of distribution. H type had seven ones such as 72H, 75H, 78H, 81H, 84H. 87H and 90H. H type had 33 observations and frequency ratio of 26.19%. Same types could be graded up and down from the reference size for the age group. And this reference size became to the starting point for developing the grading system. 2) For pattern grading of pants, fatty types, H10 type had six ones such as 80H10. 82H10, 84H10, 86H10, 88H10 and 90H10. H10 type had 28 observations and frequency ratio of 20.29%. H6 type had 6 ones such as 84H6, 86H6, 88H6, 90H6. 92H6 and 94H6. H6 type had 27 observations and frequency ratio of 19.57%. If lower body types were classified as same ones, these types could be graded up and down proportionately.

• Journal of Biosystems Engineering
• /
• v.27 no.2
• /
• pp.133-142
• /
• 2002

This is a fundamental study to develop a sensor to detect weeds in paddy field using machine vision adopted spectralphotometric technique in order to use the sensor to spread herbicide selectively. A set of spectral reflectance data was collected from dry and wet soil and leaves of rice and 6 kinds of weed to select desirable wavelengths to classify soil, rice and weeds. Stepwise variable selection method of discriminant analysis was applied to the data set and wavelengths of 680 and 802 m were selected to distinguish plants (including rice and weeds) from dry and wet soil, respectively. And wavelengths of 580 and 680 nm were selected to classify rice and weeds by the same method. Validity of the wavelengths to distinguish the plants from soil was tested by cross-validation test with built discriminant function to prove that all of soil and plants were classified correctly without any failure. Validity of the wavelengths for classification of rice and weeds was tested by the same method and the test resulted that 98% of rice and 83% of weeds were classified correctly. Feasibility of CCD color camera to detect weeds in paddy field was tested with the spectral reflectance data by the same statistical method as above. Central wavelengths of RGB frame of color camera were tried as tile effective wavelengths to distingush plants from soil and weeds from plants. The trial resulted that 100% and 94% of plants in dry soil and wet soil, respectively, were classified correctly by the central wavelength or R frame only, and 95% of rice and 85% of weeds were classified correctly by the central wavelengths of RGB frames. As a result, it was concluded that CCD color camera has good potential to be used to detect weeds in paddy field.

• Smart Media Journal
• /
• v.5 no.3
• /
• pp.54-61
• /
• 2016

According to many smart devices are development, SNS(Social Network Service) users are getting higher that is possible for real-time communicating, information sharing without limitations in distance and space. Nowadays, SNS users that based on communication and relationships, are getting uses SNS for information sharing. In this paper, we used the SNS posts for users to extract the category and information provider, how to following of recommend method. Particularly, this paper focuses on classifying the words in the text of the posts and measures the frequency using Inception-v3 model, which is one of the machine learning technique -CNN(Convolutional Neural Network) we classified image word. By classifying the category of a word in a text and image, that based on DMOZ to build the information provider DB. Comparing user categories classified in categories and posts from information provider DB. If the category is matched by measuring the degree of similarity to the information providers is classified in the category, we suggest that how to recommend method of the most similar information providers account.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.6
• /
• pp.1153-1158
• /
• 2006

The classification of the background and cell areas is very important research area because of the ambiguous boundary. In this paper, the region of cell is extracted from an image of uterine cervical cytodiagnosis using the region growing method that increases the region of interest based on similarity between pixels. Segmented image from background and cell areas is binarized using a threshold value. And then 8-directional tracking algorithm for contour lines is applied to extract the cell area. First, the extracted nucleus is transformed to RGB color that is the original image. Second, the K-means clustering algorithm is employed to classify RGB pixels to the R, G, and B channels, respectively. Third, the Hue information of nucleus is extracted from the HSI models that is the transformation of the clustering values in R, G, and B channels. The backpropagation algorithm is employed to classify and identify the normal or abnormal nucleus.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.10
• /
• pp.2180-2188
• /
• 2009

This paper is study on real-time hand gesture recognition system based on vision for intelligent robot control. We are proposed a recognition system using PCA and BP algorithm. Recognition of hand gestures consists of two steps which are preprocessing step using PCA algorithm and classification step using BP algorithm. The PCA algorithm is a technique used to reduce multidimensional data sets to lower dimensions for effective analysis. In our simulation, the PCA is applied to calculate feature projection vectors for the image of a given hand. The BP algorithm is capable of doing parallel distributed processing and expedite processing since it take parallel structure. The BP algorithm recognized in real time hand gestures by self learning of trained eigen hand gesture. The proposed PCA and BP algorithm show improvement on the recognition compared to PCA algorithm.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.8
• /
• pp.1961-1972
• /
• 2014

Recently, with the increasing of the interest in the integration of medical technology and information communication technology, researches on WBAN (Wireless Body Area Networks) that try to apply sensor network to the human body have been processed actively. The existing sensor network technology has the potential to be used in WBAN, but it has some limitations also. In particular, because the sensors are likely to communicate through each part of the body, it has a very different network environment from the sensor network that uses a free space. Therefore, researches on WBAN have a variety area of study that slightly different from the conventional sensor networks and take into account the characteristics of the body. In this study, we investigate the environmental characteristics of WBAN that are separated from the conventional sensor network, and the research trends of WBAN systematically by using the technique of SLR (Systematic Literature Review) from 2001 around when the concept of WBAN has been introduced. The investigation includes the classification of research and the researcher's features. And the survey results and the outlook for further study are summarized.

• Journal of KIISE:Databases
• /
• v.36 no.2
• /
• pp.63-72
• /
• 2009

Digital forensics refers to finding electronic evidences related to crimes. As cyber crimes are increasing daily, digital forensics for finding electronic evidences is also becoming important. At present, various aspects of digital forensics have being researched including the overall process model and analysis techniques such as network forensics, system forensics and database forensics for digital forensics. Regarding database forensics, only analysis techniques dependent on specific vendors have been suggested. And general process models and analysis techniques which can be used in various databases have not been studied. This paper proposes an integrated process model and analysis technique for database forensics. The proposed database forensics model (DFM) allows us to solve problems and analyze databases according to the situation and purpose, and to use a standard model and techniques for various database analyses. In order to test our model(DFM), we applied it to various database analyses. And we confirmed the results of our experiment that it can be applicable to acquisition in the scene as well as analysis of data relationships.

• Parasites, Hosts and Diseases
• /
• v.25 no.1
• /
• pp.37-44
• /
• 1987

As a series of systematic classification for Korean common liver fluke, Fasciola sp., karyotype was investigated by means of the modified air-drying technique and of the regular Giemsa staining. Also, C-staining method was applied for detailed karyological analysis from the germ cells of the fluke. The following is a brief summary of the leading facts gained through the experiment. 1. Korean Pasciola sp. was classified into three types based on their chromosomal complements; individuals with 20 or 30 chromosomes and with a 20/30 mosaic constitution. Worms having 30 chromosomes represent a triploid form with 3 sets of 10 basic chromosomes, while those with 20 chromosomes were diploid and mosaic individuals were 2n/3n mixoploid. 2. The frequency of the individual type calculated is as follows; 67.45% of 212 flukes examined was of diploid, 10.85%, triploid, and the rest, 21.7%, mixoploid, respectively. In many cases, two or three types were found in the peculiar bovine host while single type inhabitant was about 20% out of 52 cases. 3. The twenty chromosomes consisted of 1 pair of large metacentrics, 4 pairs of medium-sized subtelocentrics, and 5 pairs of small submetacentrics, while constitution of the thirty chromosomes was nearly interpreted as a triploid form with 3 sets of 10 basic chromosomes. The high centromeric indexes of both types are the first Pairs among all the examined, and 37.93% was of diploid and 47.93%, triploid, respectively. 4. In mixoploid individuals, constitution of the chromosomes of diploid or triploid cells was the same as that of diploid or triploid individuals. 5. All the chromosomes of the germ cells in both types showed C-band around the centromeric region and especially the chromosomes no's 3,7, and 8 showed a remarkable C-band distinguished from other chromosomes. 6. The variance for the sizes of the worms and the eggs were not parallel with three different genotypes in Korean common liver fluke.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.3231-3249
• /
• 2014

Deniable encryption, introduced in 1997 by Canetti, Dwork, Naor, and Ostrovsky, guarantees that the sender or the receiver of a secret message is able to "fake" the message encrypted in a specific ciphertext in the presence of a coercing adversary, without the adversary detecting that he was not given the real message. Sender - side deniable encryption scheme is considered to be one of the classification of deniable encryption technique which defined as resilient against coercing the sender. M. H. Ibrahim presented a sender - side deniable encryption scheme which based on public key and uncertainty of Jacobi Symbol [6]. This scheme has several problems; (1) it can't be able to derive the fake message $M_f$ that belongs to a valid message set, (2) it is not secure against Quadratic Residue Problem (QRP), and (3) the decryption process is very slow because it is based dramatically on square root computation until reach the message as a Quadratic Non Residue (QNR). The first problem is solved by J. Howlader and S. Basu's scheme [7]; they presented a sender side encryption scheme that allows the sender to present a fake message $M_f$ from a valid message set, but it still suffers from the last two mentioned problems. In this paper we present a new sender-side deniable public-key encryption scheme with fast decryption by which the sender is able to lie about the encrypted message to a coercer and hence escape coercion. While the receiver is able to decrypt for the true message, the sender has the ability to open a fake message of his choice to the coercer which, when verified, gives the same ciphertext as the true message. Compared with both Ibrahim's scheme and J. Howlader and S. Basu's scheme, our scheme enjoys nice two features which solved the mentioned problems: (1) It is semantically secure against Quadratic Residue Problem; (2) It is as fast, in the decryption process, as other schemes. Finally, applying the proposed deniable encryption, we originally give a coercion resistant internet voting model without physical assumptions.