• Title/Summary/Keyword: lightweight

Search Result 2,764, Processing Time 0.027 seconds

A Generalized Adaptive Deep Latent Factor Recommendation Model (일반화 적응 심층 잠재요인 추천모형)

  • Kim, Jeongha;Lee, Jipyeong;Jang, Seonghyun;Cho, Yoonho
    • Journal of Intelligence and Information Systems
    • /
    • v.29 no.1
    • /
    • pp.249-263
    • /
    • 2023
  • Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

Evaluation for applicability of river depth measurement method depending on vegetation effect using drone-based spatial-temporal hyperspectral image (드론기반 시공간 초분광영상을 활용한 식생유무에 따른 하천 수심산정 기법 적용성 검토)

  • Gwon, Yeonghwa;Kim, Dongsu;You, Hojun
    • Journal of Korea Water Resources Association
    • /
    • v.56 no.4
    • /
    • pp.235-243
    • /
    • 2023
  • Due to the revision of the River Act and the enactment of the Act on the Investigation, Planning, and Management of Water Resources, a regular bed change survey has become mandatory and a system is being prepared such that local governments can manage water resources in a planned manner. Since the topography of a bed cannot be measured directly, it is indirectly measured via contact-type depth measurements such as level survey or using an echo sounder, which features a low spatial resolution and does not allow continuous surveying owing to constraints in data acquisition. Therefore, a depth measurement method using remote sensing-LiDAR or hyperspectral imaging-has recently been developed, which allows a wider area survey than the contact-type method as it acquires hyperspectral images from a lightweight hyperspectral sensor mounted on a frequently operating drone and by applying the optimal bandwidth ratio search algorithm to estimate the depth. In the existing hyperspectral remote sensing technique, specific physical quantities are analyzed after matching the hyperspectral image acquired by the drone's path to the image of a surface unit. Previous studies focus primarily on the application of this technology to measure the bathymetry of sandy rivers, whereas bed materials are rarely evaluated. In this study, the existing hyperspectral image-based water depth estimation technique is applied to rivers with vegetation, whereas spatio-temporal hyperspectral imaging and cross-sectional hyperspectral imaging are performed for two cases in the same area before and after vegetation is removed. The result shows that the water depth estimation in the absence of vegetation is more accurate, and in the presence of vegetation, the water depth is estimated by recognizing the height of vegetation as the bottom. In addition, highly accurate water depth estimation is achieved not only in conventional cross-sectional hyperspectral imaging, but also in spatio-temporal hyperspectral imaging. As such, the possibility of monitoring bed fluctuations (water depth fluctuation) using spatio-temporal hyperspectral imaging is confirmed.

Research on Radiation Shielding Film for Replacement of Lead(Pb) through Roll-to-Roll Sputtering Deposition (롤투롤 스퍼터링 증착을 통한 납(Pb) 대체용 방사선 차폐필름 개발)

  • Sung-Hun Kim;Jung-Sup Byun;Young-Bin Ji
    • Journal of the Korean Society of Radiology
    • /
    • v.17 no.3
    • /
    • pp.441-447
    • /
    • 2023
  • Lead(Pb), which is currently mainly used for shielding purposes in the medical radiation, has excellent radiation shielding functions, but is continuously exposed to radiation directly or indirectly due to the harmfulness of lead itself to the human body and the inconvenience caused by its heavy weight. Research on shielding materials that are human-friendly, lightweight, and convenient to use that can block risks and replace lead is continuously being conducted. In this study, based on the commonly used polyethylene terephthalate (PET) film and the fabric material used in actual radiation protective clothing, a multi-layer thin film was realized through sputtering and vacuum deposition of bismuth, tungsten, and tin, which are metal materials that can shield radiation. Thus, a shielding film was produced and its applicability as a radiation shielding material was evaluated. The radiation shielding film was manufactured by establishing the optimized conditions for each shielding material while controlling the applied voltage, roll driving speed, and gas supply amount to manufacture the shielding film. The adhesion between the parent material and the shielding metal thin film was confirmed by Cross-cut 100/100, and the stability of the thin film was confirmed through a hot water test for 1 hour to measure the change of the thin film over time. The shielding performance of the finally realized shielding film was measured by the Korea association for radiation application (KARA), and the test conditions (inverse wide beam, tube voltage 50 kV, half layer 1.828 mmAl) were set to obtain an attenuation ratio of 16.4 (initial value 0.300 mGy/s, measured value 0.018 mGy/s) and damping ratio 4.31 (initial value 0.300 mGy/s, measured value 0.069 mGy/s) were obtained. by securing process efficiency for future commercialization, light and shielding films and fabrics were used to lay the foundation for the application of films to radiation protective clothing or construction materials with shielding functions.

Ship Detection from SAR Images Using YOLO: Model Constructions and Accuracy Characteristics According to Polarization (YOLO를 이용한 SAR 영상의 선박 객체 탐지: 편파별 모델 구성과 정확도 특성 분석)

  • Yungyo Im;Youjeong Youn;Jonggu Kang;Seoyeon Kim;Yemin Jeong;Soyeon Choi;Youngmin Seo;Yangwon Lee
    • Korean Journal of Remote Sensing
    • /
    • v.39 no.5_3
    • /
    • pp.997-1008
    • /
    • 2023
  • Ship detection at sea can be performed in various ways. In particular, satellites can provide wide-area surveillance, and Synthetic Aperture Radar (SAR) imagery can be utilized day and night and in all weather conditions. To propose an efficient ship detection method from SAR images, this study aimed to apply the You Only Look Once Version 5 (YOLOv5) model to Sentinel-1 images and to analyze the difference between individual vs. integrated models and the accuracy characteristics by polarization. YOLOv5s, which has fewer and lighter parameters, and YOLOv5x, which has more parameters but higher accuracy, were used for the performance tests (1) by dividing each polarization into HH, HV, VH, and VV, and (2) by using images from all polarizations. All four experiments showed very similar and high accuracy of 0.977 ≤ AP@0.5 ≤ 0.998. This result suggests that the polarization integration model using lightweight YOLO models can be the most effective in terms of real-time system deployment. 19,582 images were used in this experiment. However, if other SAR images,such as Capella and ICEYE, are included in addition to Sentinel-1 images, a more flexible and accurate model for ship detection can be built.