• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.221-227
• /
• 2024

In autonomous navigation systems, the need for fast and accurate image processing using deep learning and advanced sensor technologies is paramount. These systems rely heavily on the ability to process and interpret visual data swiftly and precisely to ensure safe and efficient navigation. Despite the critical importance of such capabilities, there has been a noticeable lack of research specifically focused on ship image classification for maritime applications. This gap highlights the necessity for more in-depth studies in this domain. In this paper, we aim to address this gap by presenting a comprehensive comparative study of ship image classification using two distinct neural network models: the Feedforward Neural Network (FNN) and the Convolutional Neural Network (CNN). Our study involves the application of both models to the task of classifying ship images, utilizing a dataset specifically prepared for this purpose. Through our analysis, we found that the Convolutional Neural Network demonstrates significantly more effective performance in accurately classifying ship images compared to the Feedforward Neural Network. The findings from this research are significant as they can contribute to the advancement of core source technologies for maritime autonomous navigation systems. By leveraging the superior image classification capabilities of convolutional neural networks, we can enhance the accuracy and reliability of these systems. This improvement is crucial for the development of more efficient and safer autonomous maritime operations, ultimately contributing to the broader field of autonomous transportation technology.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.8
• /
• pp.2948-2963
• /
• 2015

In the field of computer vision, visual surveillance systems have recently become an important research topic. Growth in this area is being driven by both the increase in the availability of inexpensive computing devices and image sensors as well as the general inefficiency of manual surveillance and monitoring. In particular, the ultimate goal for many visual surveillance systems is to provide automatic activity recognition for events at a given site. A higher level of understanding of these activities requires certain lower-level computer vision tasks to be performed. So in this paper, we propose an intelligent activity recognition model that uses a structure learning method and a classification method. The structure learning method is provided as a K2-learning algorithm that generates Bayesian networks of causal relationships between sensors for a given activity. The statistical characteristics of the sensor values and the topological characteristics of the generated graphs are learned for each activity, and then a neural network is designed to classify the current activity according to the features extracted from the multiple sensor values that have been collected. Finally, the proposed method is implemented and tested by using PETS2013 benchmark data.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.2
• /
• pp.110-114
• /
• 2015

This paper reports a novel object-detection technique in video sequences. The proposed algorithm consists of detection of objects in illuminated and occluded videos by using object features and a neural network technique. It consists of two functional modules: region-based object feature extraction and continuous detection of objects in video sequences with region features. This scheme is proposed as an enhancement of the Lowe's scale-invariant feature transform (SIFT) object detection method. This technique solved the high computation time problem of feature generation in the SIFT method. The improvement is achieved by region-based feature classification in the objects to be detected; optimal neural network-based feature reduction is presented in order to reduce the object region feature dataset with winner pixel estimation between the video frames of the video sequence. Simulation results show that the proposed scheme achieves better overall performance than other object detection techniques, and region-based feature detection is faster in comparison to other recent techniques.

• Journal of Digital Convergence
• /
• v.12 no.3
• /
• pp.255-261
• /
• 2014

Advances of digital device technology such as digital cameras, smart phones and tablets, provide convenience way for people to take pictures during his/her life. Photo data is being spread rapidly throughout the social network, causing the excessive amount of data available on the internet. Photo retrieval is categorized into three types, which are: keyword-based search, example-based search, visualize query-based search. The commonly used multimedia search methods which are implemented on Smart TV are adapting the previous methods that were optimized for PC environment. That causes some features of the method becoming irrelevant to be implemented on Smart TV. This paper proposes a novel Visual Query-based Photo Retrieval Method in Smart TV Environment using a motion sensing input device known as Kinect Sensor. We detected hand gestures using kinect sensor and used the information to mimic the control function of a mouse. The average precision and recall of the proposed system are 81% and 80%, respectively, with threshold value was set to 0.7.

• 전자공학회논문지 IE
• /
• v.46 no.3
• /
• pp.55-61
• /
• 2009

This paper describes an assistant system for elders who live alone. The developed system is composed of a wearable RFID system, a gateway system, and server system. The wearable RFID system is installed in glove. The wearable RFID system can be considered as a wireless sensor network which has a sink node and sensor node with a RFID reader. The sensor node can read RFID tags on the various objects used in daily living such as furniture, medicines, sugar and salt bottles, and ok. The sensor node transmits wireless packets to the sink node. The sink node sends the received packet immediately to a server system via a gateway system. The gateway provides users with audio-visual information of objects. The server system is composed of a database server and a web server. The data from each wearable RFID system is collected into a database, and then the data are processed to visualize the measurement of daily living activities of users. The processed data can be provided for someone who wants to know about user's daily living patterns in house such as family, caregivers, and medical crew.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.3
• /
• pp.1284-1292
• /
• 2015

As visual surveillance systems become more and more common in human lives, approaches based on these systems to solve security problems in practice are boosted, especially in railway applications. In this paper, we first propose a robust snag detection algorithm and then present a railway security system by using a combination of multiple sensors and the vision based snag detection algorithm. The system aims safety at several repeatedly occurred situations including slope protection, inspection of the falling-object from bridges, and the detection of snags and foreign objects on the rail. Experiments demonstrate that the snag detection is relatively robust and the system could guarantee the security of the railway through these real-time protections and detections.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.1005-1018
• /
• 2017

This paper proposes an App Visualization (AppV) based on IoT Self-diagnosis Micro Control Unit (ISMCU) for accident prevention. It collects a current status of a vehicle through a sensor, visualizes it on a smart phone and prevents vehicles from accident. The AppV consists of 5 components. First, a Sensor Layer (SL) judges noxious gas from a current vehicle and a driver's driving habit by collecting data from various sensors such as an Accelerator Position Sensor, an O2 sensor, an Oil Pressure Sensor, etc. and computing the concentration of the CO collected by a semiconductor gas sensor. Second, a Wireless Sensor Communication Layer (WSCL) supports Zigbee, Wi-Fi, and Bluetooth protocol so that it may transfer the sensor data collected in the SL to ISMCU and the data in the ISMCU to a Mobile. Third, an ISMCU integrates the transferred sensor information and transfers the integrated result to a Mobile. Fourth, a Mobile App Block Programming Tool (MABPT) is an independent App generation tool that changes to visual data just the vehicle information which drivers want from a smart phone. Fifth, an Embedded Module (EM) records the data collected through a Smart Phone real time in a Cloud Server. Therefore, because the AppV checks a vehicle' fault and bad driving habits that are not known from sensors and performs self-diagnosis through a mobile, it can reduce time and cost spending on accidents caused by a vehicle's fault and noxious gas emitted to the outside.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.2
• /
• pp.245-252
• /
• 2020

Unexpected equipment defects can cause a huge economic losses in the society at large. Although condition monitoring can provide solutions, the signal processing algorithms must be developed to predict mechanical failures using data acquired from various sensors attached to the equipment. The signal processing algorithms used in a condition monitoring requires high computing efficiency and resolution. To improve condition monitoring on a wireless sensor network(WSN), data visualization can maximize the expressions of the data characteristics. Thus, this paper proposes the extraction of visual feature from temperature data over time using condition monitoring based on a WSN to identify environmental conditions of equipment in a large-scale infrastructure. Our results show that time-frequency analysis can visually track temperature changes over time and extract the characteristics of temperature data changes.

• Journal of Sensor Science and Technology
• /
• v.21 no.1
• /
• pp.28-33
• /
• 2012

The visual sensing of human hands plays an important part in many man-machine interaction/interface systems. Most existing visionbased hand detection techniques depend on the color cues of human skin. The RGB color image from a vision sensor is often transformed to another color space as a preprocessing of hand detection because the color space transformation is assumed to increase the detection accuracy. However, the actual effect of color space transformation has not been well investigated in literature. This paper discusses a comparative evaluation of the pixel classification performance of hand skin detection in four widely used color spaces; RGB, YIQ, HSV, and normalized rgb. The experimental results indicate that using the normalized red-green color values is the most reliable under different backgrounds, lighting conditions, individuals, and hand postures. The nonlinear classification of pixel colors by the use of a multilayer neural network is also proposed to improve the detection accuracy.

• Proceedings of the KWS Conference
• /
• 2002.10a
• /
• pp.595-599
• /
• 2002

Edge welding of thin sheets is very difficult because of the fit-up problem and small weld area In laser welding, joint fit-up and penetration are critical for sound weld quality, which can be monitored by appropriate methods. Among the various monitoring systems, visual monitoring method is attractive because various kinds of weld pool information can be extracted directly. In this study, a vision sensor was adopted for the weld pool monitoring in pulsed Nd:YAG laser edge welding to monitor whether the penetration is enough and the joint fit-up is within the requirement. Pulsed Nd:YAG laser provides a series of periodic laser pulses, while the shape and brightness of the weld pool change temporally even in one pulse duration. The shutter-triggered and non-interlaced CCD camera was used to acquire a temporally changed weld pool image at the moment representing the weld status well. The information for quality monitoring can be extracted from the monitored weld pool image by an image processing algorithm. Weld pool image contains not only the information about the joint fit-up, but the penetration. The information about the joint fit-up can be extracted from the weld pool shape, and that about a penetration from the brightness. Weld pool parameters that represent the characteristics of the weld pool were selected based on the geometrical appearance and brightness profile. In order to achieve accurate prediction of the weld penetration, which is nonlinear model, neural network with the selected weld pool parameters was applied.