Detection of Wildfire Burned Areas in California Using Deep Learning and Landsat 8 Images (딥러닝과 Landsat 8 영상을 이용한 캘리포니아 산불 피해지 탐지)
-
- Korean Journal of Remote Sensing
- /
- v.39 no.6_1
- /
- pp.1413-1425
- /
- 2023
The increasing frequency of wildfires due to climate change is causing extreme loss of life and property. They cause loss of vegetation and affect ecosystem changes depending on their intensity and occurrence. Ecosystem changes, in turn, affect wildfire occurrence, causing secondary damage. Thus, accurate estimation of the areas affected by wildfires is fundamental. Satellite remote sensing is used for forest fire detection because it can rapidly acquire topographic and meteorological information about the affected area after forest fires. In addition, deep learning algorithms such as convolutional neural networks (CNN) and transformer models show high performance for more accurate monitoring of fire-burnt regions. To date, the application of deep learning models has been limited, and there is a scarcity of reports providing quantitative performance evaluations for practical field utilization. Hence, this study emphasizes a comparative analysis, exploring performance enhancements achieved through both model selection and data design. This study examined deep learning models for detecting wildfire-damaged areas using Landsat 8 satellite images in California. Also, we conducted a comprehensive comparison and analysis of the detection performance of multiple models, such as U-Net and High-Resolution Network-Object Contextual Representation (HRNet-OCR). Wildfire-related spectral indices such as normalized difference vegetation index (NDVI) and normalized burn ratio (NBR) were used as input channels for the deep learning models to reflect the degree of vegetation cover and surface moisture content. As a result, the mean intersection over union (mIoU) was 0.831 for U-Net and 0.848 for HRNet-OCR, showing high segmentation performance. The inclusion of spectral indices alongside the base wavelength bands resulted in increased metric values for all combinations, affirming that the augmentation of input data with spectral indices contributes to the refinement of pixels. This study can be applied to other satellite images to build a recovery strategy for fire-burnt areas.
As wildfires are difficult to predict, real-time monitoring is crucial for a timely response. Geostationary satellite images are very useful for active fire detection because they can monitor a vast area with high temporal resolution (e.g., 2 min). Existing satellite-based active fire detection algorithms detect thermal outliers using threshold values based on the statistical analysis of brightness temperature. However, the difficulty in establishing suitable thresholds for such threshold-based methods hinders their ability to detect fires with low intensity and achieve generalized performance. In light of these challenges, machine learning has emerged as a potential-solution. Until now, relatively simple techniques such as random forest, Vanilla convolutional neural network (CNN), and U-net have been applied for active fire detection. Therefore, this study proposed an active fire detection algorithm using state-of-the-art (SOTA) deep learning techniques using data from the Advanced Himawari Imager and evaluated it over East Asia and Australia. The SOTA model was developed by applying EfficientNet and lion optimizer, and the results were compared with the model using the Vanilla CNN structure. EfficientNet outperformed CNN with F1-scores of 0.88 and 0.83 in East Asia and Australia, respectively. The performance was better after using weighted loss, equal sampling, and image augmentation techniques to fix data imbalance issues compared to before the techniques were used, resulting in F1-scores of 0.92 in East Asia and 0.84 in Australia. It is anticipated that timely responses facilitated by the SOTA deep learning-based approach for active fire detection will effectively mitigate the damage caused by wildfires.
Due to recent global climate change, the scale of flood damage is increasing as rainfall is concentrated and its intensity increases. Rain on a scale that has not been observed in the past may fall, and long-term rainy seasons that have not been recorded may occur. These damages are also concentrated in ASEAN countries, and many people in ASEAN countries are affected, along with frequent occurrences of flooding due to typhoons and torrential rains. In particular, the Bandung region which is located in the Upper Chitarum River basin in Indonesia has topographical characteristics in the form of a basin, making it very vulnerable to flooding. Accordingly, through the Official Development Assistance (ODA), a flood forecasting and warning system was established for the Upper Citarium River basin in 2017 and is currently in operation. Nevertheless, the Upper Citarium River basin is still exposed to the risk of human and property damage in the event of a flood, so efforts to reduce damage through fast and accurate flood forecasting are continuously needed. Therefore, in this study an artificial intelligence-based river flood water level forecasting model for Dayeu Kolot as a target station was developed by using 10-minute hydrological data from 4 rainfall stations and 1 water level station. Using 10-minute hydrological observation data from 6 stations from January 2017 to January 2021, learning, verification, and testing were performed for lead time such as 0.5, 1, 2, 3, 4, 5 and 6 hour and LSTM was applied as an artificial intelligence algorithm. As a result of the study, good results were shown in model fit and error for all lead times, and as a result of reviewing the prediction accuracy according to the learning dataset conditions, it is expected to be used to build an efficient artificial intelligence-based model as it secures prediction accuracy similar to that of using all observation stations even when there are few reference stations.
This study attempted to identify the relationship between the investment determinants of accelerators and investment performance through empirical analysis. Through literature review, four dimensions and 12 measurement items were extracted for investment determinants, which are independent variables, and investment performance was adjusted to the cumulative amount of subsequent investment based on previous studies. Performance data from 594 companies selected by TIPS from 2017 to 2019, which are relatively reliable and easy to secure data, were collected, and the subsequent investment cumulative attraction amount, which is a dependent variable, was hypothesized through multiple regression analysis three years after the investment. As a result of the study, 'industrial experience years' in the characteristics of founders, 'market size', 'market growth', 'competitive strength', and 'number of patents' in the characteristics of products and services had a significant positive (+) effect. The impact of independent variables on dependent variables was most influenced by the competitive strength of market characteristics, followed by the number of years of industrial experience, the number of patents, the size of the market, and market growth. This was different from the results of previous studies conducted mainly on qualitative research methods, and in most previous studies, the characteristics of founders were the most important, but the empirical analysis results were market characteristics. As a sub-factor, the intensity of competition, which was the subordinate to the importance of previous studies, had the greatest influence in empirical analysis. The academic significance of this study is that it presented a specific methodology to collect and build 594 empirical samples in the absence of empirical research on accelerator investment determinants, and created an opportunity to expand the theoretical discussion of investment determinants through causal research. In practice, the information asymmetry and uncertainty of startups that accelerators have can help them make effective investment decisions by establishing a systematic model of experience-dependent investment determinants.
The wall shear stress in the vicinity of end-to end anastomoses under steady flow conditions was measured using a flush-mounted hot-film anemometer(FMHFA) probe. The experimental measurements were in good agreement with numerical results except in flow with low Reynolds numbers. The wall shear stress increased proximal to the anastomosis in flow from the Penrose tubing (simulating an artery) to the PTFE: graft. In flow from the PTFE graft to the Penrose tubing, low wall shear stress was observed distal to the anastomosis. Abnormal distributions of wall shear stress in the vicinity of the anastomosis, resulting from the compliance mismatch between the graft and the host artery, might be an important factor of ANFH formation and the graft failure. The present study suggests a correlation between regions of the low wall shear stress and the development of anastomotic neointimal fibrous hyperplasia(ANPH) in end-to-end anastomoses. 30523 T00401030523 ^x Air pressure decay(APD) rate and ultrafiltration rate(UFR) tests were performed on new and saline rinsed dialyzers as well as those roused in patients several times. C-DAK 4000 (Cordis Dow) and CF IS-11 (Baxter Travenol) reused dialyzers obtained from the dialysis clinic were used in the present study. The new dialyzers exhibited a relatively flat APD, whereas saline rinsed and reused dialyzers showed considerable amount of decay. C-DAH dialyzers had a larger APD(11.70
The wall shear stress in the vicinity of end-to end anastomoses under steady flow conditions was measured using a flush-mounted hot-film anemometer(FMHFA) probe. The experimental measurements were in good agreement with numerical results except in flow with low Reynolds numbers. The wall shear stress increased proximal to the anastomosis in flow from the Penrose tubing (simulating an artery) to the PTFE: graft. In flow from the PTFE graft to the Penrose tubing, low wall shear stress was observed distal to the anastomosis. Abnormal distributions of wall shear stress in the vicinity of the anastomosis, resulting from the compliance mismatch between the graft and the host artery, might be an important factor of ANFH formation and the graft failure. The present study suggests a correlation between regions of the low wall shear stress and the development of anastomotic neointimal fibrous hyperplasia(ANPH) in end-to-end anastomoses. 30523 T00401030523 ^x Air pressure decay(APD) rate and ultrafiltration rate(UFR) tests were performed on new and saline rinsed dialyzers as well as those roused in patients several times. C-DAK 4000 (Cordis Dow) and CF IS-11 (Baxter Travenol) reused dialyzers obtained from the dialysis clinic were used in the present study. The new dialyzers exhibited a relatively flat APD, whereas saline rinsed and reused dialyzers showed considerable amount of decay. C-DAH dialyzers had a larger APD(11.70
Out of a total of 58 species of helminthic and protozoan parasitic infections in Korea, so far 38 species were reported as intestinal parasites of man. Quite a few species of the intestinal parasitic infections have long been prevalent throughout the country and this has been a significant public health problem. In this paper, current status and transition of the intestinal parasitic infections in the past years were presented. Chronological reviewing of data show background and prospects of change in the prevalence of infections. In the national prevalence survey on parasitic infections conducted once every five years since 1971, stool examinations were done by both cellophane thick smear and brine flotation techniques. Every egg positive case of Ascaris lumbricoides and Clonorchis sinensis was further examined by Stoll's dilution egg counting technique. In 1981, perianal swab using adhesive cellulose tape was added for Enterobius infection. For protozoan cyst examination conducted by province and city in '81, fecal specimens were fixed in SAF solution and examined by the formalin-ether concentration technique. High prevalence of parasitic infection in ana before the 1960s can be easily understood from the data given by the Ministry of Health and Social Affairs in 1967. From a parasite control point of view, the 1960s was the preparatory period particularly for control of soil-transmitted helminthiasis. Several organizations which have contributed to parasite control were founded in this period and the prevention law of parasitic infections was passed in '66. In the '70s, overall prevalence rates of the common intestinal parasitic infections, which were highly prevalent in the past were turned into reduction phase for the first time. The '80s can be said to be an active control period of parasitic infections. Intestinal helminths According to the reports of the national prevalence survey on parasitic infections, the prevalence of helminthic infections was 84.3%(number of person examined:24,887) in '71, 63.2%(27,178) in '76, and 41.1%(35,018) in '81. By area, the prevalence rate in '81 was 35.1%(20,569) in urban areas and 49.6%(14,447) in rural areas. Intestinal nematodes Ascaris lumbricoides The prevalence of Ascaris infection has decreased significantly in recent years. Among students, the prevalence was 55.4%, in '69 and decreased to 4.7% in '83. In the national prevalence surveys, the prevalence decreased to 13.0% in '81 from 54.9% in '71. By area, the prevalence decreased to 8.5% in '81 from 46.4% in '71 in urban area and 19.4% in '81 from 59.6% in '71 in rural area. By age, the prevalence has become in recent years relatively even in all age groups, although higher prevalence used to be seen in young age groups of around 10 years old, particulary in the highly prevalent rural areas. By sex, the prevalence was higher in the female than in the male. Unfertilized egg positive rates among the ascariasis cases increased gradually up to 55.4% on the average in '81. The intensity of the infection was also significantly decreased. Trichuris trichiura Trichuris infection had also decreased to 23.4% in '81 from 65.4% in '71. By area, the decreasing tendency of the prevalence became faster in urban areas than in the rural areas. The prevalence in urban and rural areas in '71 was 69.7%, and 63.1% respectively and decreased to 19.5% and 29.0% respectively in '81. By age, the prevalence reached a peak at the 10-14 age group and showed relatively even distribution throughout all age groups. By sex, the prevalence was close in young age groups, but in the 30s or over age group, especially in rural area, the prevalence was significantly higher in the female than in the male. The prevalence has much fluctuated depending in the area. The prevalence in rural areas surveyed in the '80s shows a range between 20.9% and 73.7% by locality. It is anticipated that the prevalence of Trichuris infection will drop more rapidly, when mass treatment is conducted. Hookworms Hookworm infection by mostly Ancylostoma duodenale and a few by Necator americanus has decreased to a negligible levels in recent years. In the national prevalence surveys, the prevalence was 10.7% in '71, 2.2% in '76, and 0.5% in '81. The prevalence was higher in rural areas than in urban areas. Wide application of multi-specific anthelminthics in the ascariasis control programmes conducted in the past decade appear to have been effective against hookworm infection. Trichostrongylus orientalis As in the case with hookworm infection, the prevalence of Trichostrongylus infection has reached a negligible levels. In the national prevalence surveys, the prevalence was 7.7% in '71, 1.0% in '76 and 0.2% in '81. Enterobius vermicularis In the national prevalence survey in '81, the egg positive rate was 12.0%. Higher prevalence is expected when examined repeatedly. The prevalence rate was 10.3% in urban area and 14.6% in rural area. In recent surveys conducted in rural areas among schoolchildren, the prevalence was 32.4% in Gimhae Gun in '82 and 64.1% in Yeongyang Gun in '83. By age, the egg positive rate was higher in young age groups of around 10 and sharply decreased in age groups of around 20 and then somewhat increased again in middle age groups. By sex, the prevalence was higher in the female than in the male. Strongyloides stercoralis Strongyloides stercoralis infection has rarely been found in Korea. Three cases were reported in 1914. And 0.1-0.5% were found infected out of 2,642 persons examined at the prisoner-of-war camp on Geojedo in 1956. One case was reported in '54 and '82, respectively. Anisakis spp. No systematic survey has been conducted for anisakiasis In Korea. So far, only several cases have been found 1 case in Seoul in '71, 5 cases in Busan in '81 and 1 case in Busan in '84. Intestinal trematodes Metagonimus yokogawai In the national prevalence survey conducted in 1981, the egg positive rate was 1.24% on the average. High endemic areas are located in the southwestern part of Korea. The prevalence in Hadong Gun was 29.1% on the average in '79. In a survey conducted in 76, the prevalence was 44.0% in Gwangyang, 55.0% in Gogseung and 29.0% in Gurye. The infection is closely correlated with raw sweetfish consumption in these areas. Other intestinal trematodes A human case of Heterophyes heterophyes was reported in 1914. Several species were reported in the '80s : 17 cases of Fibricola seoulensis, 9 cases of Pygidiopsis summa, 8 cases of Heterophyes heterophyes nocens, 1case of Heterophyopsis continua, 2 cases of Stellantchasmus falcatus, 1 case of Stictodora sp., 1 case of Echinostoma hortense, and 4 cases of Echinochasmus japonicus. As the intermediate hosts, snakes and frogs play a role for F.seoulensis and fish for the rest of the species. Intestinal cestodes Taenia saginata and T. solium Egg positive rates in the national prevalence survey were 0.7% in '76 and 1.1% in '81. The prevalence in '81 was 0.6% in urban area and 1.8% in rural area. The proglottid positive rate in Jeju Do was 19.2% on the average. On Udo, Jeju Do in 1983, the egg positive rate among the inhabitants was 2.9%. Hymenolepis nana In the national prevalence survey, egg positive rates were 0.6% in '76 and 0.4% in '81. No difference was seen in the prevalence by area and sex. Hymenolepis diminuta Infected cases were reported : 3 in '64 and I in '66. Egg positive rate in '81 was 0.01% in the national prevalence survey. Diphylobothrium latum So far, about 30 cases have been reported. The cases have been reported more frequently in recent surveys. Mesocestoides sp. A case was reported from a hospitalized patient in Seoul in '67. Spirometra erinacei Two cases were reported in '84 following reidentification of the adult worms collected in '74. Intestinal protozoa Out of a total of 23 species of human protozoan infections in Korea, 13 species were reported as intestinal protozoa : Entamoeba histolytica, E coli, Endolimax nana, Iodamoeba b
Purpose: We aimed to investigate the efficacy and side effects of
During April 1993 to November 1994, cations, anions, and conductivity were analyzed to examine how summer monsoon influences the ionic content of Taechung Reservoir, Korea. Interannual variability of ionic content reflected hydrological characteristics between the two years(high-flood year in 1993 vs. draught year in 1994). Cations, anions and conductivity were lowest during peak inflow in 1993 and highest during a drought in 1994. Floods in 1993 markedly decreased total salinity as a result of reduced Ca
contrasting monsoon between 1993 and 1994 produced an interannual difference in hydrology. Theoretical water residence time (TWRT) in monsoon 1993 averaged 27 d, which was>3 months shorter compared to the TWRT in monsoon 1994. A dominant physical process influencing thermal stratification, water movement, and mixing regime was an interflow current in 1993. During summer 1993, river water plunged to mid-lake (location 27 km) and passed through the 10