• Journal of the Korea Society of Computer and Information
• /
• v.29 no.7
• /
• pp.41-51
• /
• 2024

In this paper, we explore the enhancement of target detection accuracy in the guided weapon using deep learning object detection on infrared (IR) images. Due to the characteristics of IR images being influenced by factors such as time and temperature, it's crucial to ensure a consistent representation of object features in various environments when training the model. A simple way to address this is by emphasizing the features of target objects and reducing noise within the infrared images through appropriate pre-processing techniques. However, in previous studies, there has not been sufficient discussion on pre-processing methods in learning deep learning models based on infrared images. In this paper, we aim to investigate the impact of image pre-processing techniques on infrared image-based training for object detection. To achieve this, we analyze the pre-processing results on infrared images that utilized global or local information from the video and the image. In addition, in order to confirm the impact of images converted by each pre-processing technique on object detector training, we learn the YOLOX target detector for images processed by various pre-processing methods and analyze them. In particular, the results of the experiments using the CLAHE (Contrast Limited Adaptive Histogram Equalization) shows the highest detection accuracy with a mean average precision (mAP) of 81.9%.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.7
• /
• pp.641-646
• /
• 2011

A low cost gait analysis system, which can measure stride length, walking speed, and ground reaction force, is proposed. A gait analysis system is used for medical evaluation of patients and rehabilitation assistance. Low cost cameras are attached to a shoe and movement of a shoe is estimated using infrared LED landmarks. Ground reaction force is measured from pressure sensors, which are installed inside a shoe. Through experiments, it is shown that the proposed system can be used to obtain stride length, walking speed, and ground reaction force.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1521-1521
• /
• 2001

The objective of the research being presented was to construct and calibrate a spectrometer for the analysis of single kernels of corn. In light of the difficulties associated with capturing the spatial variability in composition of corn kernels by single-beam spectrometry, a hyperspectral imaging spectrometer was constructed with the intention that it would be used to analyze single kernels of corn for the prediction of moisture and oil content. The spectrometer operated in the range of 750- 1090 nanometers. After evaluating four methods of standardizing the output from the spectrometer, calibrations were made to predict whole-kernel moisture and oil content from the hyperspectral image data. A genetic algorithm was employed to reduce the number of wavelengths imaged and to optimize the calibrations. The final standard errors of prediction during cross-validation (SEPCV) were 1.22% and 1.25% for moisture and oil content, respectively. It was determined, by analysis of variance, that the accuracy and precision of single-kernel corn analysis by hyperspectral imaging is superior to the single kernel reference chemistry method (as tested).

• Journal of Korean Institute of Fire Investigation
• /
• v.11 no.1
• /
• pp.37-40
• /
• 2008

This paper deals with damage patterns of cabinet panel for low voltage deteriorated by flame. In order to analyze damage patterns, we used Metallurgical Microscope, x-ray system, and Fourier Transform Infrared spectroscopy. Firstly, Metallurgical microscope was used for analysis of electrical causes, such as electric short and overload Secondly, X-ray system was used for analysis of internal characteristics of circuit breakers. Lastly, Fourier Transform Infrared spectroscopy was used for analysis of damage direction by flame.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2008.04a
• /
• pp.269-272
• /
• 2008

This paper deals with damage patterns of cabinet panel for low voltage deteriorated by flame. In order to analyze damage patterns, we used Metallurgical Microscope, x-ray system, and Fourier Transform Infrared spectroscopy. Firstly, Metallurgical microscope was used for analysis of electrical causes, such as electric short and overload. Secondly, X-ray system was used for analysis of internal characteristics of circuit breakers. Lastly, Fourier Transform Infrared spectroscopy was used for analysis of damage direction by flame. The following results were obtained.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.2
• /
• pp.139-143
• /
• 2011

The problem of delamination of composite tubes by impact has been acknowledged in aerospace and automobile industry. Non-destructive testing(NDT) methods in composite material structure are important to evaluate reliability of composite structure. There are many kinds of NDT methods which can detect the inside defect of the composite material such as Infrared Thermography(IRT). Infrared thermal imaging of object is different from that of a defect, in heated composite tubes with an internal defect, and then location and size of a defect can be measured by the analysis of thermal imaging pattern. In this study, Lock-in Infrared thermography detect internal defects of Impacted composite tubes by the inspection of infrared lay radiated from the surface of composite tubes.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1254-1254
• /
• 2001

Near infrared spectroscopy (NIRS) was applied to determination of the lipid content of compost during compost fermentation of tofu(soybean-curd) refuse. The reflected rays in the wavelength range between 800 and 2500 nm were measured at 2 nm intervals. The absorption of lipid observed at 4 wavelengths, 1208, 1712, 2312 and 2352 nm on the second derivative spectra. To formulate a calibration equation, a multiple linear regression analysis was carried out between the near-infrared spectral data and on the lipid content in the calibration sample set (sample number, n=60) obtained using a Soxhlet extraction method. The calibration equation for prediction of lipid, the value of the multiple correlation coefficient (R) was 0.975 when using the wavelengths of 1208 and 1712nm. To validate the calibration equation obtained, the lipid content in the validation sample set (n=35) not used for formulating the calibration equation were calculated using the calibration equations, and compared with the values obtained using the Soxhlet extraction method. Good agreement were observed between the results of the Soxhlet extraction method and those values of the NIRS method. The simple correlation coefficient (r) and standard error of prediction (SEP) were 0.964 and 0.815 %, respectively. Then, the NIRS method was applied to a compost fermentation in which the time course the lipid content were measured and good results were obtained. The study indicates that NIRS is a useful method for process management of the compost fermentation of tofu refuse.

• Journal of Institute of Control, Robotics and Systems
• /
• v.17 no.12
• /
• pp.1203-1209
• /
• 2011

In this paper, the degradation evaluation method of VLC (Visible Light Communication) wireless module after high dose rate gamma-ray irradiation using the thermal infrared camera is proposed. First, the heating characteristics of the active devices embedded in the VLC wireless module during the condition of normal operation is monitored by thermal infrared camera. By the image processing technique, the trends of the intensity of the heat emitted by the active devices are calculated and stored. The feature of the blob area including the area of the active devices in the thermal infrared image is extracted and stored. The feature used in this paper is the mean value of the gray levels in the blob area. The same VLC module has been gamma irradiated at the dose rate of about 4.0 kGy/h during 72 hours up to a total dose of 288 kGy. And then, the heating characteristics of the active devices embedded in the VLC wireless module after high dose gamma ray irradiation is observed by thermal infrared camera. The high dose gamma-ray induced degradation of the active devices embedded in the VLC module was evaluated by comparing the mean value of the blob area to the one of the same blob area of the VLC module before the gamma ray irradiation.

• 전자공학회논문지 IE
• /
• v.48 no.2
• /
• pp.32-39
• /
• 2011

The output image of the uncooled thermal infrared camera is difficult the identification of target because of the limited dynamic range and the various noises. Retinex algorithm based on the theory of the human visual perception is known to be effective contrast enhancement technique. However, the image quality is insufficient when it is adopted to the narrow dynamic range image as the infrared image. In this paper, we propose the revised retinex algorithm to enhance the contrast of the infrared image. To improve the contrast enhancement performance, we designed the new dynamic range compression function instead of log function. To reduce the noise and compensate the loss of edge, we added the contrast compensation procedure in the MSR image generation process. According to the output picture comparing and numerical analysis, the proposed algorithm shows the better contrast enhancement performance and the more suitable method for the infrared image enhancement.

• Elastomers and Composites
• /
• v.44 no.1
• /
• pp.47-54
• /
• 2009

EPDM(Ethylene-propylene-diene-terpolymer) compound reinforced with carbon black having four different particle size, acetylene black(thermal conductivity carbon black), and silica were manufactured by internal mix and open mill. To investigate the effect of particle size of filler and filler type on far-infrared vulcanization, intermal temperature of compound, degree of curing, infrared spectroscopy, and thermal analysis were measured. The thermal conductivity of far-infrared vulcanized EPDM compound increased with increasing particle size of carbon filler, but hot air vulcanized EPDM compound is not affected by particle size. The thermal conductivity was increased in the order of carbon black < silica < acetylene black(thermal conductivity carbon black).