• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1433-1449
• /
• 2023

In recent years, the rapid development of social networks has led to a rapid increase in the amount of information available on the Internet, which contains a large amount of sensitive information related to pornography, politics, and terrorism. In the aspect of sensitive image detection, the existing machine learning algorithms are confronted with problems such as large model size, long training time, and slow detection speed when auditing and supervising. In order to detect sensitive images more accurately and quickly, this paper proposes a multiclassification sensitive image detection method based on lightweight Convolutional Neural Network. On the basis of the EfficientNet model, this method combines the Ghost Module idea of the GhostNet model and adds the SE channel attention mechanism in the Ghost Module for feature extraction training. The experimental results on the sensitive image data set constructed in this paper show that the accuracy of the proposed method in sensitive information detection is 94.46% higher than that of the similar methods. Then, the model is pruned through an ablation experiment, and the activation function is replaced by Hard-Swish, which reduces the parameters of the original model by 54.67%. Under the condition of ensuring accuracy, the detection time of a single image is reduced from 8.88ms to 6.37ms. The results of the experiment demonstrate that the method put forward has successfully enhanced the precision of identifying multi-class sensitive images, significantly decreased the number of parameters in the model, and achieved higher accuracy than comparable algorithms while using a more lightweight model design.

• Korean Journal Plant Pathology
• /
• v.10 no.4
• /
• pp.297-304
• /
• 1994

This study was carried out to develop bioassay for detection of pesticide residues in agricultural products by using the soil microbial isolates sensitive to pesticides. One hundred bacterial isolates and eighty five fungal isolates were obtained from soil and their sensitivity to 10 ppm of several pesticides was examined in vitro. Five bacterial isolates and three fungal isolates were found sensitive to organochloride fungicide and two fungal isolates sensitive to organocopper fungicide. Among these isolates, B46, B93 and F67 were tested to find out the difference in sensitivity according to the methods of fungicide treatment. All of the isolates were found sensitive to 10 ppm of organochloride fungicides mixed directly in PDA. But they were found insensitive to the fungicide mixed in PDA after filtering through membrane filter. In case of organocopper fungicide, the isolates were found sensitive only when it was treated in PDA. And their sensitivity showed difference among various kinds of organochloride fungicides. B46 and B93 were employed to check the possibility as the agent for detection of the pesticidal residues in twenty eight agricultural products including rice. It was found that all samples had not residues because the samples did not inhibit the growth of isolates. When organochloride fungicides were applied to the above products, it was possible to detect the residues in fruits and vegetables at the concentration of 10 ppm, but not in starch-rich grains. B46 and B93 were identified as Bacillus sp. according to their bacterial characteristics in culture.

• Journal of Microbiology and Biotechnology
• /
• v.18 no.11
• /
• pp.1858-1861
• /
• 2008

A PCR-enzyme-linked immunosorbent assay (PCR-ELISA) was developed for the rapid and sensitive detection of L. monocytogenes. PCR primers generating a 132-bp amplicon and a capture probe able to hybridize to the PCR amplicon were designed based on the L. monocytogenes-specific hly gene encoding listeriolysin. The detection limit of PCR-ELISA for L. monocytogenes was determined to be as low as 10 cells per PCR reaction, and this level of detection was achieved within 5 h. These results indicate that the PCR-ELISA provides a valuable tool for the rapid and sensitive detection of L. monocytogenes for the ready-to-eat food industry.

• Korean Journal of Veterinary Pathology
• /
• v.6 no.1
• /
• pp.1-5
• /
• 2002

A1celaphine herpesvirus 1 (AHV-1) is a causative agent of malignant catarrhal fever which is a fatal and a lymphoproliferative syndrome. AHV-1 is a gamma herpesvirus, which induces frequent latent infection and often difficult to detect its antigens or specific nucleic acids because of its low viral copies in the infected tissues. A new method, in situ PCR, is developed for the detection of AHV-1 nucleic acid in this study. Target sequences of AHV-1 open reading frame 50 gene were detected within AHV-1 infected MDBK cells. As compare with other molecular biological methods for the detection of AHV-1, in situ PCR was found to be more sensitive than in situ hybridization and to be less sensitive than nested PCR. However, nested PCR cannot afford to observe and differentiate AHV-1 infected cells. In situ PCR amplifies a target sequence within cells that can be visualized microscopically with increased sensitivity compared to detection by in situ hybridization. In situ PCR has wide applications for sensitive localization of low copy AHV-1 viral sequences within cells to investigate the role of viruses in a variety of clinical conditions and also provide the rapid, sensitive, and specific detection of AHV-1 infection.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.5
• /
• pp.545-551
• /
• 2005

The main objective of fraud detection is to minimize costs or losses that are incurred due to fraudulent transactions. Because of the problem's nature such as highly skewed, overlapping class distribution and non-uniform misclassification costs, it is, however, practically difficult to generate a classifier that is near-optimal in terms of classification costs at a desired operating range of rejection rates. This paper defines a performance measure that reflects classifier's costs at a specific operating range and offers a cost-sensitive learning approach that enables us to train classifiers suitable for real-world credit card fraud detection by directly optimizing the performance measure with evolutionary programming. The experimental results demonstrate that the proposed approach provides an effective way of training cost-sensitive classifiers for successful fraud detection, compared to other training methods.

• Bulletin of the Korean Chemical Society
• /
• v.26 no.6
• /
• pp.979-982
• /
• 2005

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.3
• /
• pp.271-280
• /
• 2004

Assessment of image registration for Pressure Sensitive Paint (PSP) was performed. A 16 bit camera and LED lamp were used with Uni-FIB paint (ISSI). Because of model displacement and deformation at 'wind-on' condition, a large error of the intensity ratio was induced between 'wind-on' and' wind-off images. To correct the error, many kinds of image registrations were tested. At first, control points were marked on the model surface to find the coefficients of polynomial transform functions between the 'wind-off' 'wind-on' images. The 2nd-order polynomial function was sufficient for representing the model displacement and deformation. An automatic detection scheme was introduced to find the exact coordinates of the control points. The present automatic detection algorithm showed more accurate and user-friendly than the manual detection algorithm. Since the coordinates of transformed pixel were not integer, five interpolation methods were applied to get the exact pixel intensity after transforming the 'wind-on' image. Among these methods, the cubic convolution interpolation scheme gave the best result.

• Asian-Australasian Journal of Animal Sciences
• /
• v.17 no.3
• /
• pp.423-427
• /
• 2004

Seventy-six animals including cattle, sheep, horses, 6 species of zoo animals were employed for collection of fresh feces in order to detect verotoxigenic Esherichia coli (VTEC) by safe, quick and sensitive PCR-based molecular methods. Bacterial cell disruption with bead-beating followed by bacterial DNA purification with hydroxyapatide chromatography and gel filtration allowed DNA preparation from animal feces with high recovery and purity. A mountain goat was firstly shown by PCR and sequencing to shed verotoxin 2 gene (vt2) that was used to generate vt2 probe and second primer set for nested PCR to attempt more sensitive detection. Most sensitive nested PCR revealed that 45% of tested cattle and 47% of tested zoo animals were VTEC-positive, while least sensitive normal PCR detected VTEC from none of these animals except a mountain goat. Moderately sensitive detection by PCR in combination with hybridization suggested that the VTEC density varied between the VTEC-positive cattle.

• Proceedings of the KIEE Conference
• /
• 2005.10b
• /
• pp.367-369
• /
• 2005

In this paper, we present a new method for the detection of spiculation on digital mammograms. Traditional methods have defects; sensitive to noise, fixed size processing, and long processing time, however, the proposed method has merits; not sensitive to noise, adaptive size processing, and fast processing time. Experimental results show that the spiculation detection performance of the proposed method is improved much compared to the other methods.

• Food Science and Biotechnology
• /
• v.16 no.3
• /
• pp.337-342
• /
• 2007

Conventional methods for pathogen detection and identification are labor-intensive and take days to complete. Biosensors have shown great potential for the rapid detection of foodborne pathogens. Fiber-optic biosensors have been used to rapidly detect pathogens because they can be very sensitive and are simple to operate. However, many fiber-optic biosensors rely on manual sensor handling and the sandwich assay, which require more effort and are less sensitive. To increase the simplicity of operation and detection sensitivity, a binding inhibition assay method for detecting Listeria monocytogenes in food samples was developed using an automated, fiber-optic-based immunosensor: RAPTOR (Research International, Monroe, WA, USA). For the assay, fiber-optic biosensors were developed by the immobilization of Listeria antibodies on polystyrene fiber waveguides through a biotin-avidin reaction. Developed fiber-optic biosensors were incorporated into the RAPTOR to evaluate the detection of L. monocytogenes in frankfurter samples. The binding inhibition method combined with RAPTOR was sensitive enough to detect L. monocytogenes ($5.4{\times}10^7\;CFU/mL$) in a frankfurter sample.