• MALSORI
• /
• no.63
• /
• pp.139-151
• /
• 2007

As the Internet becomes prevalent in our lives, harmful contents, such as phornographic videos, have been increasing on the Internet, which has become a very serious problem. To prevent such an event, there are many filtering systems mainly based on the keyword-or image-based methods. The main purpose of this paper is to devise a system that classifies pornographic videos based on the audio information. We use the mel-cepstrum modulation energy (MCME) which is a modulation energy calculated on the time trajectory of the mel-frequency cepstral coefficients (MFCC) as well as the MFCC as the feature vector. For the classifier, we use the well-known Gaussian mixture model (GMM). The experimental results showed that the proposed system effectively classified 98.3% of pornographic data and 99.8% of non-pornographic data. We expect the proposed method can be applied to the more accurate classification system which uses both video and audio information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.4 no.5
• /
• pp.910-924
• /
• 2010

As Internet usage has increased, the risk of adolescents being exposed to adult content and harmful information on the Internet has also risen. To help prevent adolescents accessing this content, a novel detection method for adult images is proposed. The proposed method involves three steps. First, the Image Of Interest (IOI) is extracted from the image background. Second, the IOI is distinguished from the segmented image using a novel weighting mask, and it is determined to be acceptable or unacceptable. Finally, the features (color and texture) of the IOI or original image are compared to a critical value; if they exceed that value then the image is deemed to be an adult image. A Receiver Operating Characteristic (ROC) curve analysis was performed to define this optimal critical value. And, the textural features are identified using a gray level co-occurrence matrix. The proposed method increased the precision level of detection by applying a novel weighting mask and a receiver operating characteristic curve. To demonstrate the effectiveness of the proposed method, 2850 adult and non-adult images were used for experimentation.

• Korean Journal of Community Nutrition
• /
• v.3 no.1
• /
• pp.21-33
• /
• 1998

The purpose of this study was to investigate the weight control attempts and related factors among 220 female college students in Seoul. Factors examined included body image, body satisfaction, interest in weight control, beliefs related to weight control, social norms, social expectations regarding subjects' body size and weight change. Data were analyzed using t-test and chi-square test(at $\alpha$=0.05). The average height and weight of subjects were 160.7cm, 52.2kg respectively. BMI and body fat(%) were 20.2 and 25.2%, respectively. One hundred seventy two students(78.2%) had attempted to control their weight, mainly to lose weight, and were categorized as the weight control attempt group. Subjects used diet modification as well as unhealthy method to control weight. 48.2% in the attempt group were underweight or normal weight(by BMI), suggesting that their weight control attempts were was unnecessary. Students in the attempt group described their own size as heavier than those in the no-attempt group(p<0.01) or other girls of their age(p<0.05) ; more dissatisfied with their body size(p<0.01), and showed more interest in weight control(p<0.01). Several differences in beliefs were also noted(p<0.001). The attempt group believed less strongly in the harmful effects or difficulties in weight control and believed more strongly in the advantages of weight control. With respect to social factors, the attempt group perceived that their family and friends wanted them to be smaller than they were(p<0.001), perceived that significant others felt that they should lose weight(p<0.05). These results suggest that educational programs for college female students should start focusing on the harmful effects of excessive dieting and information about desirable weight control methods. Students should be helped to have a correct body image. In addition, educational programs should incorporate strategies to change beliefs regarding weight control, as well as modifying social expectations from significant others.

• Journal of Environmental Impact Assessment
• /
• v.4 no.2
• /
• pp.59-70
• /
• 1995

The solution-state of a reclaimed waste would be known to the method in using an analysis for seepage. But it is not the best method in the huge landfill reclaimed all kinds of the waste at random. Especially in case of the landfill called the Nan Gi-do located along the Han-river, it is difficult to judge the generative seepage to be flowed in to the Han-river. So to plan the effective stabilization on a landfill, it is very useful survey method using the Landsat TM image. Operating a heat-distribution analysis with the Landsat TM image, in case of a landfill not having definite data, we would assume the reclaimed sections of the waste to judge a solution-speed late comparatively such as a industry waste or a harmful waste through the heat change.

• Journal of Biosystems Engineering
• /
• v.24 no.5
• /
• pp.439-444
• /
• 1999

Automated greenhouse production systems often require crop growth monitoring involving accurate quantification of plant physiological properties. Conventional methods are usually burdensome, inaccurate, and harmful to crops. A thermal image analysis system can accomplish rapid and accurate measurements of physiological-property changes of stressed crops. In this research a thermal imaging system was used to measure the leaf-temperature changes of several crops according to water deficit. Thermal images were obtained from lettuce, cucumber, pepper, and chinese cabbage plants. Results showed that there were significant differences in the temperature of stressed plants and non-stressed plants. The temperature differences between these two group of plants were 0.7 to 3$^{\circ}C$ according to species.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.63-64
• /
• 2014

본 논문에서는 성인영상에서 모자이크 검출 기법을 제안한다. 먼저 성인영상에 가장 많이 사용되는 격자형 모자이크 검출을 위하여 입력영상에서 에지를 검출한 후, 에지영상을 이용하여 모자이크 후보영역을 검출한다. 모자이크 후보영역은 모자이크맵을 생성하고 이진화하여 후보영역을 산출한다. 이렇게 산출된 후보영역내의 모자이크맵의 값이 임계치 이상인 영역을 최종 모자이크 영역으로 검출한다. 실험결과 모자이크가 없는 영상에서 약간의 오검출이 있었으나, 전체적으로 우수한 성능을 보였다.

• Asia Marketing Journal
• /
• v.11 no.4
• /
• pp.49-67
• /
• 2010

Previous researches have proven that corporate social responsibility(adhere CSR) is positively related to corporate performance. But Most of CSR related researches have several limitations. One of limitations is that those researches treated CSR as unidimensional construct. Almost researchers in the area of CSR concepts insisted that CSR is consist of multi dimensions. Carroll's four dimensions of CSR have been utilized by numerous academicians. Carroll asserted that CSR is composed of four dimensions: economic, legal, ethical, and philanthropic responsibility. But Carroll's dimensions were revised as three dimensions by Schwartz and Carroll, because ethical and philanthropic responsibility are not mutually exclusive. If CSR construct is composed of multiple dimensions, a message related to one of dimensions changes beliefs or evaluations about other dimensions that are not mentioned in the message. This phenomenon is called as "spillover effect". According to Ahluwalia, Unnava, and Burnkrant, negative information spills over to attributes that are associated with the target attributes but not mentioned in the message. Like this, this preponderant effect of negative information over positive information has been termed the "negativity effect". In this paper, authors try to prove the spillover effect and negativity effect among Schwartz and Carroll's three dimensions(economic, legal, and ethical responsibility) of CSR. The results of this study show that messages related to legal and ethical responsibility cause spillover effect and influence consumers' evaluation to other dimensions. Moreover, when negativity effect is added on spillover effect, spillover effect is more increased. It means that negative messages related to legal and ethical responsibility is more harmful to corporate image than negative message related to economic responsibility. The results of this study will help companies to manage corporate image using CSR messages as marketing communication tools. Companies should manage messages related to legal and ethical responsibility for more efficiently managing corporate image. Specially, because negative messages related to legal and ethical responsibility are more harmful to corporate image, companies must take care not to spread out negative message related to legal and ethical responsibility. Finally, we discuss the implications of the findings and limitations.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.05a
• /
• pp.687-690
• /
• 2004

본 논문에서는 음란유해콘텐츠를 기술적으로 차단하기 위해 신호처리 기법에 기반한 방법을 제안하고자 한다. 지금까지의 기술적인 방법 등이 목록기반과 단어기반에 기초한 방법이기 때문에 음란콘텐츠가 제대로 차단이 되지 않은 경우가 많았다. 이를 위해 본 논문에서는 신호처리에 기반하여 음란 유해 콘텐츠를 차단하는 방법을 제안하고자 한다. 끝으로 실험에 의해 제안한 방법의 유용성을 입증하고자 한다.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2000.11c
• /
• pp.593-599
• /
• 2000

Accurate quantification of plant physiological properties is often necessary for optimal control of an automated greenhouse production system. Conventional crop growth monitoring systems are usually burdensome, inaccurate, and harmful to crops. A thermal image analysis system was used to accomplish rapid and accurate measurements of physiological-property changes of water-stressed crops. Thermal images were obtained from several species of plants that were placed in a growth chamber. Analyzing the images provided the pattern of temperature changes in a leaf and the amount of differences in the temperature of stressed plants and non-stressed plants.

• International Journal of Computer Science & Network Security
• /
• v.24 no.4
• /
• pp.67-76
• /
• 2024

Diabetic retinopathy is a threatening complication of diabetes, caused by damaged blood vessels of light sensitive areas of retina. DR leads to total or partial blindness if left untreated. DR does not give any symptoms at early stages so earlier detection of DR is a big challenge for proper treatment of diseases. With advancement of technology various computer-aided diagnostic programs using image processing and machine learning approaches are designed for early detection of DR so that proper treatment can be provided to the patients for preventing its harmful effects. Now a day machine learning techniques are widely applied for image processing. These techniques also provide amazing result in this field also. In this paper we discuss various machine learning and deep learning based techniques developed for automatic detection of Diabetic Retinopathy.