• The Journal of the Korea institute of electronic communication sciences
• /
• 제17권1호
• /
• pp.71-76
• /
• 2022

Since brightness is proportional to the operating current, a method of connecting several LEDs in series and driving with a constant current source is widely used for driving circuits of LED lights. Because several LEDs are connected in series, if some LEDs open due to a fault, the current path is broken and all other LEDs connected in series are turned off. In this paper, we designed a circuit to solve this problem by connecting a Zener diode having a breakdown voltage of about 0.4V higher than the LED operating voltage in parallel with each LED to create a current bypass in case of LED failure. Through simulations and experiments, it was confirmed that the current of the Zener diode hardly flows when the LED is operating normally, and that the Zener diode stably operates as a current bypass when the LED fails.

• Journal of IKEEE
• /
• 제26권3호
• /
• pp.468-474
• /
• 2022

In this paper, we propose a method to implement a real-time fire alarm system using deep learning. The deep learning image dataset for fire alarms acquired 1,500 sheets through the Internet. If various images acquired in a daily environment are learned as they are, there is a disadvantage that the learning accuracy is not high. In this paper, we propose a fire image data expansion method to improve learning accuracy. The data augmentation method learned a total of 2,100 sheets by adding 600 pieces of learning data using brightness control, blurring, and flame photo synthesis. The expanded data using the flame image synthesis method had a great influence on the accuracy improvement. A real-time fire detection system is a system that detects fires by applying deep learning to image data and transmits notifications to users. An app was developed to detect fires by analyzing images in real time using a model custom-learned from the YOLO V4 TINY model suitable for the Edge AI system and to inform users of the results. Approximately 10% accuracy improvement can be obtained compared to conventional methods when using the proposed data.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제26권10호
• /
• pp.1453-1461
• /
• 2022

The accuracy of peripheral object recognition algorithms using 3D data sensors such as LiDAR in autonomous vehicles has been increasing through many studies, but this requires high performance hardware and complex structures. This object recognition algorithm acts as a large load on the main processor of an autonomous vehicle that requires performing and managing many processors while driving. To reduce this load and simultaneously exploit the advantages of 3D sensor data, we propose 2D data-based recognition using the ROI generated by extracting physical properties from 3D sensor data. In the environment where the brightness value was reduced by 50% in the basic image, it showed 5.3% higher accuracy and 28.57% lower performance time than the existing 2D-based model. Instead of having a 2.46 percent lower accuracy than the 3D-based model in the base image, it has a 6.25 percent reduction in performance time.

• Journal of Broadcast Engineering
• /
• 제26권6호
• /
• pp.809-812
• /
• 2021

Air bubbles which may be generated during the PCB coating process can be a major cause of malfunction. so it is necessary to detect the bubbles in advance. In previous studies, candidates for bubbles were extracted using the brightness characteristics of bubbles, and the candidates were verified using CNN(Convolutional Neural Networks). In this paper, we propose a bubble detection method using a transfer learning-based CNN model. The VGGNet is adopted and sigmoid is used as a classification layer, and the last convolutional layer and classification layer are trained together when transfer learning is applied. The performance of the proposed method is F1-score 0.9044, which shows an improvement of about 0.17 compared to the previous study.

• Journal of the Institute of Convergence Signal Processing
• /
• 제24권1호
• /
• pp.34-41
• /
• 2023

In recent years, as the range of human activities has increased, the introduction of alien species has become frequent. Among them, raccoons have been designated as harmful animals since 2020. Raccoons are similar in size and shape to raccoon dogs, so they generally need to be distinguished in capturing them. To solve this problem, we use VGG19, ResNet152V2, InceptionV3, InceptionResNet and NASNet, which are CNN deep learning models specialized for image classification. The parameters to be used for learning are pre-trained with a large amount of data, ImageNet. In order to classify the raccoon and raccoon dog datasets as outward features of animals, the image was converted to grayscale and brightness was normalized. Augmentation methods were applied using left and right inversion, rotation, scaling, and shift to create sufficient data for transfer learning. The FCL consists of 1 layer for the non-augmented dataset while 4 layers for the augmented dataset. Comparing the accuracy of various augmented datasets, the performance increased as more augmentation methods were applied.

• Journal of Internet of Things and Convergence
• /
• 제10권2호
• /
• pp.51-60
• /
• 2024

Aiming at the problem of low contrast and difficult to recognize video images in low-light environment, we propose an adaptive contrast compensation enhancement algorithm based on human visual perception. First of all, the video image characteristic factors in low-light environment are extracted: AL (average luminance), ABWF (average bandwidth factor), and the mathematical model of human visual CRC(contrast resolution compensation) is established according to the difference of the original image's grayscale/chromaticity level, and the proportion of the three primary colors of the true color is compensated by the integral, respectively. Then, when the degree of compensation is lower than the bright vision precisely distinguishable difference, the compensation threshold is set to linearly compensate the bright vision to the full bandwidth. Finally, the automatic optimization model of the compensation ratio coefficient is established by combining the subjective image quality evaluation and the image characteristic factor. The experimental test results show that the video image adaptive enhancement algorithm has good enhancement effect, good real-time performance, can effectively mine the dark vision information, and can be widely used in different scenes.

• Journal of the Korea Computer Graphics Society
• /
• 제30권1호
• /
• pp.11-17
• /
• 2024

Lung cancer segmentation in chest CT images is challenging due to the varying sizes of tumors and the presence of surrounding structures with similar intensity values. To address these issues, we propose a lung cancer segmentation network that incorporates deep supervision and utilizes UNet3+ as the backbone. Additionally, we propose a hybrid lesion focal loss function comprising three components: pixel-based, region-based, and shape-based, which allows us to focus on the smaller tumor regions relative to the background and consider shape information for handling ambiguous boundaries. We validate our proposed method through comparative experiments with UNet and UNet3+ and demonstrate that our proposed method achieves superior performance in terms of Dice Similarity Coefficient (DSC) for tumors of all sizes.

• The Journal of the Korea institute of electronic communication sciences
• /
• 제19권1호
• /
• pp.165-172
• /
• 2024

In this paper, we designed a smart home system that can be used intuitively and easily in everyday life, such as sending text messages to users, providing various information and scheduling using smart AI, and providing lighting and atmosphere suitable for the atmosphere in situations such as listening to music using neopixels, as well as using ESP32, RFID, and Google Cloude Console using raspberry pie. As a result of the experiment, it was confirmed that security characters were normally sent to users when RFID was recognized on ESP32 connected to Wi-Fi even if the power was reconnected, and smart AI using Neopixel lighting, Raspberry Pie, and voice recognition, which calculated frequency, also changed the recognition rate over distance, but it worked.

• Journal of the Korea Computer Graphics Society
• /
• 제30권4호
• /
• pp.11-18
• /
• 2024

In partial nephrectomy for kidney cancer treatment, accurate segmentation of the kidney tumor is crucial for surgical planning, as it provides essential information on the precise size and location of the tumor. However, it is challenging due to the tumor's similar intensity to surrounding organs and the variability in its location and size across patients. In this study, we propose a hybrid network that integrates a convolutional neural network and a transformer to capture both local and global features, aiming to improve the segmentation performance of kidney tumors. We validated our method through comparative experiments with UNETR++, outperforming it with a Dice Similarity Coefficient (DSC) of 78.54% and a precision of 85.0 7%. Moreover, in the analysis by tumor size, our method demonstrated improvements by reducing over-segmentation and outlier cases observed in UNETR++.

• Journal of the Korean earth science society
• /
• 제25권8호
• /
• pp.754-763
• /
• 2004

This study was aimed to examine the Landsat Enhanced Thematic Mapper Plus (ETM+) index, which matches well with the field survey data in the wildfire area of Gangneung, Gangweon Province, Korea. In the wildfire area NDVI (Normalized Difference Vegetation Index), SAVI (Soil Adjusted Vegetation Index), and Tasseled Cap Transformation Index (Brightness, Wetness, Greenness) were compared with field survey data. NDVI and SAVI were very useful in detecting the difference between the wildfire and non-wildfire area, but not so in classify the soil types in the wildfire area. The soil plane based on the Tasseled Cap Transformation showed a better result in classifying the soil types in the wildfire areas than NDVI and SAVI, and corresponded well with field survey data. Using a linear function based on greenness and wetness in the Tasseled Cap Transformation is expected to provide a more efficient and quicker method to classify wildfire areas.