• KIPS Transactions on Computer and Communication Systems
• /
• v.7 no.6
• /
• pp.155-164
• /
• 2018

Many malicious programs have been compressed or encrypted using various commercial packers to prevent reverse engineering, So malicious code analysts must decompress or decrypt them first. The OEP (Original Entry Point) is the address of the first instruction executed after returning the encrypted or compressed executable file back to the original binary state. Several unpackers, including PinDemonium, execute the packed file and keep tracks of the addresses until the OEP appears and find the OEP among the addresses. However, instead of finding exact one OEP, unpackers provide a relatively large set of OEP candidates and sometimes OEP is missing among candidates. In other words, existing unpackers have difficulty in finding the correct OEP. We have developed new tool which provides fewer OEP candidate sets by adding two methods based on the property of the OEP. In this paper, we propose two methods to provide fewer OEP candidate sets by using the property that the function call sequence and parameters are same between packed program and original program. First way is based on a function call. Programs written in the C/C++ language are compiled to translate languages into binary code. Compiler-specific system functions are added to the compiled program. After examining these functions, we have added a method that we suggest to PinDemonium to detect the unpacking work by matching the patterns of system functions that are called in packed programs and unpacked programs. Second way is based on parameters. The parameters include not only the user-entered inputs, but also the system inputs. We have added a method that we suggest to PinDemonium to find the OEP using the system parameters of a particular function in stack memory. OEP detection experiments were performed on sample programs packed by 16 commercial packers. We can reduce the OEP candidate by more than 40% on average compared to PinDemonium except 2 commercial packers which are can not be executed due to the anti-debugging technique.

• Journal of Broadcast Engineering
• /
• v.19 no.5
• /
• pp.578-589
• /
• 2014

Window image is displayed through a monitor screen when we execute the application programs on the computer. This includes webpage, video player and a number of applications. The webpage delivers a variety of information by various types in comparison with other application. Unlike a natural image captured from a camera, the window image like a webpage includes diverse components such as text, logo, icon, subimage and so on. Each component delivers various types of information to users. However, the components with different characteristic need to be divided locally, because text and image are served by various type. In this paper, we divide window images into many sub blocks, and classify each divided region into background, text and subimage. The detected subimages can be applied into 2D-to-3D conversion, image retrieval, image browsing and so forth. There are many subimage classification methods. In this paper, we utilize AdaBoost for verifying that the machine learning-based algorithm can be efficient for subimage detection. In the experiment, we showed that the subimage detection ratio is 93.4 % and false alarm is 13 %.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.2
• /
• pp.62-72
• /
• 2017

Spatio-temporal projection of evapotranspiration over croplands would be useful for assessment of climate change impact and development of adaptation strategies in agriculture. Potential evapotranspiration (PET) and dryness index (DI) during rice growing seasons were calculated using climate change scenario data provided by the National Institute of Meteorological Research (NIMR). A data processing tool for gridded climate data files, readGrADSWrapper, was used to calculate PET and DI during the current (1986-2005) and future (2006-2100) periods. Scripts were written to implement the formulas of PET and DI in R, which is an open source statistical data analysis tool. Evapotranspiration in rice fields ($PET_{Rice}$) was also determined using R scripts. The Spatio-temporal patterns of PET differed by regions in Korean Peninsula under current and future climate conditions. Overall, PET and $PET_{Rice}$ tended to increase throughout the $21^{st}$ century. Those results suggested that region-specific water resource managements would be needed to minimize the risk of water loss in the regions where considerable increases in PET would occur under the future climate conditions. For example, a number of provinces classified as a humid region were projected to become a sub-humid region in the future. The Spatio-temporal assessment of water resources based on PET and DI would help the development of climate change adaptation strategies for rice production in the 21st century. In addition, the studies on climate change impact would be facilitated using specialized data tools, e.g., readGrADSWrapper, for geospatial analysis of climate data.

• Journal of Korea Multimedia Society
• /
• v.13 no.9
• /
• pp.1337-1347
• /
• 2010

Fingerprint recognition is a biometric technology to identify individual by using fingerprint features such ridges and valleys. Most fingerprint systems perform the recognition based on minutiae points after acquiring a fingerprint image from contact type sensor. They have an advantage of acquiring a clear image of uniform size by touching finger on the sensor. However, they have the problems of the image quality can be reduced in case of severely dry or wet finger due to the variations of touching pressure and latent fingerprint on the sensor. To solve these problems, the contactless capturing devices for a fingerprint image was introduced in previous works. However, the accuracy of detecting minutiae points and recognition performance are reduced due to the degradation of image quality by the illumination variation. So, this paper proposes a new LDP-based fingerprint recognition method. It can effectively extract fingerprint patterns of iterative ridges and valleys. After producing histograms of the binary codes which are extracted by the LDP method, chi square distance between the enrolled and input feature histograms is calculated. The calculated chi square distance is used as the score of fingerprint recognition. As the experimental results, the EER of the proposed approach is reduced by 0.521% in comparison with that of the previous LBP-based fingerprint recognition approach.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.18 no.4
• /
• pp.45-71
• /
• 1991

The landscape systems in Korean island settlements can be recognized as results of ingabitants' ecological adptation to the isolated environment with the limited natural resources. Both the fishery dominant industry in island society and ecological nature of its environments seem to have influenced on inhabitants' environmental cognition as well as the physical landscape of island villages such as its location, spatial pattern in each village, housing form and so on. This study was done mainly by both refering to the related documents and direct observations in case study areas, and results of the study can be summarized as follows. 1. In general, the landscape of an individual island seems to take more innate characteristics of island's own, corresponding to the degree of isolation from mainland. That is, while the landscape of island in neighboring waters takes both inland-like and island-innate landscape character at the same time, the one in the open sea far from land takes more innate landscape character of all island's own in the aspects of village location, land use and housing density etc. 2. The convex landform of most islands brings about more centrifugal village allocation than centripetal allocation in most inland villages. And thus most villages in each island face extremely diverse directions different from the south facing preference in most inland rural villages. 3. Most island villages tend to be located along the ecologically transitional strip between land and sea, so called 'line of life', rather than between hilly slope and flat land as being in most inland village locations. So they are located with marine ecology bounded fishing ground ahead and land ecology bounded agricultural site at the back of them. 4. The settlement pattern of the island fishing villages shows more compact spatial structure than that of inland agricultural villages, due to the absolute limits of usable land resources and the adaptation to the marine environment with severe sea winds and waves or for the easy accessability to the fishing grounds. And also the managerial patterns of public owned sea weed catching ground, which take each family as the unit of usership rather than an individual, seem to make the villagescape more compact and the size of Individual residence smaller than that of inland agricultural village. 5. The folk shrine('Dand') systems, in persrective of villagescape, represent innate environmental cognition of island inhabitants above all other cultural landscape elements in the island. Usually the kinds and the meanings of island's communal shrine and its allocative patternsin island villagescape are composed of set with binary opposition, for example 'Upper shrine(representing 'earth', 'mountain' or 'fire')' and 'Lower Shrine(representing 'sea', 'dragon' or 'water') are those. They are usually located at contrary positions in villagescape each other. That is, they are located at 'the virtical center or visual terminus(Upper shrine at hillside behind the village)' and 'the border or entrance(Lower Shrine at seashore in front of the village)'. Each of these shirines' divinity coincides with each subsystem of island's natural eco-system(earth sphere vs marine sphere) and they also contribute to ecological conservation, bonded with the 'Sacred Forest(usually with another function of windbreak)' or 'Sacred Natural Fountain' nearby them, which are representatives of island's natural resources.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.5
• /
• pp.41-49
• /
• 2010

Being popularized the use of portable entertainment/information devices, the demand on flash memory has been also increased radically. In general, flash memory reveals various error patterns by the devices it is mounted, and thus the memory makers are trying to minimize error ratio in the final process through not only the electric test but also the data integrity test under the same condition as real application devices. This process is called an application-level memory test. Though currently various flash memory testing devices have been used in the production lines, most of the works related to memory test depend on the sensual abilities of human testers. In case of testing the flash memory for MP3 devices, the human testers are checking if the memory has some errors by hearing the audio played on the memory testing device. The memory testing process like this has become a bottleneck in the flash memory production line. In this paper, we propose an audio comparison technique to support the efficient flash memory test for MP3 devices. The technique proposed in this paper compares the variance change rate between the source binary file and the decoded analog signal and checks automatically if the memory errors are occurred or not.

• Polymer(Korea)
• /
• v.36 no.4
• /
• pp.542-548
• /
• 2012

Morphological characteristics and formation of symmetric polystyrene-block-poly(1,4-butadiene) copolymer (PS-b-PBD) in thin films upon solvent-annealing were investigated by using atomic force microscopy (AFM). The thin films solvent-annealed in cyclohexane revealed the perforated lamellae of poly(1,4-butadiene) in the matrix of polystyrene while those solvent-annealed in n-hexane exhibited highly disordered patterns. Interestingly, when the thin films of PS-b-PBD were solvent-annealed with binary mixtures of cyclohexane and n-hexane, the morphological transition from the perforated lameallae to the perpendicularly-oriented lamellae of poly(1,4-butadiene) could be induced by changing the mixing ratio of both solvents. We also demonstrated that after microdomians of poly(1,4-butadiene) were successfully degraded by UV-$O_3$, linear poly(dimethyl siloxane) chains were back-filled into the etched regions of the thin film and then converted to silica nano-objects by oxygen plasma treatments.

• The KIPS Transactions:PartD
• /
• v.9D no.4
• /
• pp.555-560
• /
• 2002

Subsequence matching, which consists of index searching and post-processing steps, is an operation that finds those subsequences whose changing patterns are similar to that of a given query sequence from a time-series database. This paper discusses optimization of post-processing for subsequence matching. The common problem occurred in post-processing of previous methods is to compare the candidate subsequence with the query sequence for discarding false alarms whenever each candidate subsequence appears during index searching. This makes a sequence containing candidate subsequences to be accessed multiple times from disk, and also have a candidate subsequence to be compared with the query sequence multiple times. These redundancies cause the performance of subsequence matching to degrade seriously. In this paper, we propose a new optimal method for resolving the problem. The proposed method stores ail the candidate subsequences returned by index searching into a binary search tree, and performs post-processing in a batch fashion after finishing the index searching. By this method, we are able to completely eliminate the redundancies mentioned above. For verifying the performance improvement effect of the proposed method, we perform extensive experiments using a real-life stock data set. The results reveal that the proposed method achieves 55 times to 156 times speedup over the previous methods.

• Korean Journal of Soil Science and Fertilizer
• /
• v.39 no.1
• /
• pp.21-28
• /
• 2006

We investigated the mobilization of Cd, Pb, and Cr in two different soils in response to sorption capacities and competition for available sorption site while they moved under saturated water conditions. Two soil samples that were clay and sandy clay loam were collected within 20 cm from the upland surface. To do this, we used three different systems of heavy metal combinations such as single, binary, and ternary as solution phase. And then we observed the breakthrough curve (BTC) and elution as a function of pore volume by applying heavy metal solution and displacing K solution until these curves reached to maximum and minimum. The results showed that BTC and elution curves were not symmetric and it required more pore volumes with increasing species of heavy metals in solution phase, as well as longer tailings. Compared the areas over and under BTC and elution curve, relatively small amount of heavy metal was displaced by K even though there were differences in electronegativity among heavy metals. Conclusively, we assumed that heavy metals transport in soil could be influenced by soil physical nonequilibrium and chemical equilibrium in solution as far as there were more than two species of heavy metals existed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.651-658
• /
• 2021

Because bolts with anti-loosening coatings are used mainly for joining safety-related components in automobiles, accurate automatic screening of these coatings is essential to detect defects efficiently. The performance of the convolutional neural network (CNN) used in a previous study [Identification of bolt coating defects using CNN and Grad-CAM] increased with increasing number of data for the analysis of image patterns and characteristics. On the other hand, obtaining the necessary amount of data for coated bolts is difficult, making training time-consuming. In this paper, resorting to the same VGG16 model as in a previous study, transfer learning was applied to decrease the training time and achieve the same or better accuracy with fewer data. The classifier was trained, considering the number of training data for this study and its similarity with ImageNet data. In conjunction with the fully connected layer, the highest accuracy was achieved (95%). To enhance the performance further, the last convolution layer and the classifier were fine-tuned, which resulted in a 2% increase in accuracy (97%). This shows that the learning time can be reduced by transfer learning and fine-tuning while maintaining a high screening accuracy.