• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.260-270
• /
• 2023

We introduce a novel Visual Inertial Odometry (VIO) algorithm designed to improve the performance of thermal-inertial odometry. Thermal infrared image, though advantageous for feature extraction in low-light conditions, typically suffers from a high noise level and significant information loss during the 8-bit conversion. Our algorithm overcomes these limitations by approximating a 14-bit raw pixel histogram into a Gaussian mixture model. The conversion method effectively emphasizes image regions where texture for visual tracking is abundant while reduces unnecessary background information. We incorporate the robust learning-based feature extraction and matching methods, SuperPoint and SuperGlue, and zero velocity detection module to further reduce the uncertainty of visual odometry. Tested across various datasets, the proposed algorithm shows improved performance compared to other state-of-the-art VIO algorithms, paving the way for robust thermal-inertial odometry.

• Journal of Sensor Science and Technology
• /
• v.27 no.5
• /
• pp.300-305
• /
• 2018

The uncooled microbolometer thermal sensor for low cost and mass volume was designed to target the new infrared market that includes smart device, automotive, energy management, and so on. The microbolometer sensor features 80x60 pixels low-resolution format and enables the use of wafer-level vacuum packaging (WLVP) technology. Read-out IC (ROIC) implements infrared signal detection and offset correction for fixed pattern noise (FPN) using an internal digital to analog convertor (DAC) value control function. A reliable WLVP thermal sensor was obtained with the design of lid wafer, the formation of Au80%wtSn20% eutectic solder, outgassing control and wafer to wafer bonding condition. The measurement of thermal conductance enables us to inspect the internal atmosphere condition of WLVP microbolometer sensor. The difference between the measurement value and design one is $3.6{\times}10-9$ [W/K] which indicates that thermal loss is mainly on account of floating legs. The mean time to failure (MTTF) of a WLVP thermal sensor is estimated to be about 10.2 years with a confidence level of 95 %. Reliability tests such as high temperature/low temperature, bump, vibration, etc. were also conducted. Devices were found to work properly after accelerated stress tests. A thermal camera with visible camera was developed. The thermal camera is available for non-contact temperature measurement providing an image that merged the thermal image and the visible image.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.37 no.1
• /
• pp.64-75
• /
• 2013

This study suggests basic data on optimum thermal insulation for spring wear through an investigation of subjective thermal sensation, self-health image and actual wearing conditions. A survey of university students using a self-administered questionnaire was conducted to collect data on subjective thermal sensation, self-health image, wearing conditions, demographics and physical characteristics. The variable of wearing conditions was measured as the response to the clothing they were wearing. Garment items (26 types for males and 41 types for females) were suggested and the items worn by the students were converted into the thermal insulation values for clothing. The main results are as follows. As for the body type perception, males perceived themselves as not fat while females perceived themselves as not thin. As for the health perception, males perceived themselves healthier than females. As for the climate adaptability perception, females were more sensitive to cold than males. The average thermal insulation of clothing was 0.97clo (0.34-1.95clo) with higher insulation for males than females. Students were more sensitive to the cold when their BMI was lower, their body surface area per body weight was larger, and the more they perceived themselves as not healthy. There was a significant correlation between the self-health image of sensitiveness to cold and the thermal insulation of clothing. The results were synthetically discussed in terms of environmental physiology.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.65 no.2
• /
• pp.130-135
• /
• 2016

Gait recognition has advantage of non-contact type recognition. But It has disadvantage of low recognition rate when the pedestrian silhouette is changed due to bag or coat. In this paper, we proposed new method using combination of gait energy image feature and thermal face image feature. First, we extracted a face image which has optimal focusing value using human body rate and Tenengrad algorithm. Second step, we extracted features from gait energy image and thermal face image using linear discriminant analysis. Third, calculate euclidean distance between train data and test data, and optimize weights using genetic algorithm. Finally, we compute classification using nearest neighbor classification algorithm. So the proposed method shows a better result than the conventional method.

• Journal of the Korean Society for Precision Engineering
• /
• v.25 no.1
• /
• pp.92-98
• /
• 2008

The flame image observation and analysis has been investigated for combustion monitoring and management of the pulverized coal firing for thermal power plant, especially for lower nitrogen oxide generation and safer operation. We aimed at obtaining the relationship between burner flame image information and emissions of nitrogen oxide and unburned carbon in furnace utilizing the flame image processing methods, by which we quantitatively determine the conditions of combustion on the individual homers. Its feasibility test was undertaken with Samchonpo thermal power plant #4 unit which has 24 burners, through which the system was observed to be effective for evaluating the combustion conditions and continuous monitoring to prevent future loss of ignition.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.25 no.4
• /
• pp.465-471
• /
• 2015

Objectives: An objective of this study is to apply a thermal image camera which shows various color according to temperature of indoor surface for estimating concentration of airborne fungi. Materials and Methods: While wall temperature were monitored by applying the thermal image camera, airborne bacteria as well as air temperature and relative humidity have been measured in lecture room and toilet of university for seven months. Results: Based on the results obtained from this study, the ranges of temperature and airborne fungi concentration were $20{\sim}24^{\circ}C$ and $20{\sim}400cfu/m^3 $ for red image, $17.5{\sim}20^{\circ}C$ and $35{\sim}150cfu/m^3$ for orange image, $15.5{\sim}17.5^{\circ}C$ and $25{\sim}650cfu/m^3$ for sky-blue image, and $13.5{\sim}15.5^{\circ}C$ and $50{\sim}200cfu/m^3$ for blue image, respectively. The color of indoor surface taken shot by thermal image camera showed consistent trend with temperature of indoor surface. There is, however, little correlation between color of indoor surface and airborne fungi concentration(p>0.05). Among environmental factors, relative humidity in indoor air showed a significant relationship with airborne fungi concentration(p<0.05). Conclusions: The more measurement data for proving statistically an association between color of indoor surface and airborne fungi concentration should be provided to easily estimate indoor level of airborne fungi.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.57-59
• /
• 2005

In our study, the automatic tracking algorithm of a human face is proposed by utilizing the thermal properties and 2nd momented geometrical feature of an infrared image. First, the facial candidates are estimated by restricting the certain range of thermal values, and the spurious blobs cleaning algorithm is applied to track the refined facial region in an infrared image.

• Journal of the Korea Safety Management & Science
• /
• v.22 no.1
• /
• pp.1-8
• /
• 2020

Maintenance of power distribution facilities is a significant subject in the power supplies. Fault caused by deterioration in power distribution facilities may damage the entire power distribution system. However, current methods of diagnosing power distribution facilities have been manually diagnosed by the human inspector, resulting in continuous pole accidents. In order to improve the existing diagnostic methods, a thermal image analysis model is proposed in this work. Using a thermal image technique in diagnosis field is emerging in the various engineering field due to its non-contact, safe, and highly reliable energy detection technology. Deep learning object detection algorithms are trained with thermal images of a power distribution facility in order to automatically analyze its irregular energy status, hereby efficiently preventing fault of the system. The detected object is diagnosed through a thermal intensity area analysis. The proposed model in this work resulted 82% of accuracy of detecting an actual distribution system by analyzing more than 16,000 images of its thermal images.

• Fire Science and Engineering
• /
• v.26 no.2
• /
• pp.69-74
• /
• 2012

Thermal imaging technology based on IR sensor with high transmittance through the fire smoke is considered as one of the advanced technology for the fire service. The present study has been performed to investigate the image saturation characteristics with thermal condition of target and background and evaluate the performance of image quality based on the contrast transfer function (CTF). For the present testing conditions, TIC using BST sensor did not show the image saturation and the image quality based on the CTF was proportional to the temperature difference between target and background. This study can be utilized as preliminary study to improve reliability and technical development of TIC.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2003.07a
• /
• pp.191-194
• /
• 2003

Through using aspheric lens can result in advantages such as an improved optical transfer function, a reduced distortion path or the realization of special image field curvatures. Using the diamond turning method for generating aspherics, the company claim to be able to generate surfaces with a form error of less than $0.33\;{\mu}\;m$ and a surface roughness of less than $0.025\;{\mu}\;m$. In this paper, we are manufacturing thermal image aspheric lens. Thermal image system is electro-optical imaging device which can make visible the difference of infrared energy naturally emitted by objected. In the result of aspherical surface, the form accuracy of about $0.24\;{\mu}\;m$ P-V was obtained and the surface roughness Ra $0.004\;{\mu}\;m$. Also, a brief review of Ultra-precision system Korea photonics technology institute(KOPTI) is present in this paper.