Tue Huu Huynh

Bio: Tue Huu Huynh is an academic researcher from International University, Cambodia. The author has contributed to research in topics: Intelligent word recognition & Mobile phone. The author has an hindex of 4, co-authored 7 publications receiving 42 citations.

PLC Impulsive Noise in Industrial Zone: Measurement and Characterization

Abstract:  Abstract—Noise in PLC is of relatively complex structure of which the most important component is the asynchronous impulsive noise. Based on measurements realized in an industrial zone, this paper shows the heavy tail phenomenon observed in experimental measures. Consequently, the alpha stable model is the most natural to be used in describing the statistics of PLC impulsive noise in industrial zones. Furthermore, the inter-arrival time is of Pareto distribution while the duration of impulses follows a mixed exponential distribution. All parameters of the alpha stable model as well as the Pareto parameters and the mixed exponents are statistically estimated. The variation of these parameters with time is also discussed.

Flexible and efficient wireless sensor networks for detecting rainfall-induced landslides

Abstract: The effect of climate change and human activities leads to a series of dangerous phenomena, such as landslides and flood. In such a context, building a system to monitor environmental hazards is seriously needed. Some studies propose to use wireless sensor network (WSN) technique for landslide monitoring systems. Two important factors for these systems are the flexibility and the energy management. This paper focuses on the development of a flexible and efficient WSN for detecting rainfall-induced landslides. A flexible switching between star and tree topologies is used to adapt to weather conditions in order to maximize the reliability of the transmission. Moreover, the power management is designed concurrently with the weather condition to improve both the operation reliability and the power efficiency. The performance of our designed WSN is evaluated by both simulation and outdoor experiment. Detailed calculation of the power budget is presented thoughtfully to confirm the effectiveness of the proposed scheme.

A study about the reconstruction of remote, low resolution mobile captured text images for OCR

Abstract: With the burst of low-cost and versatile mobile phones, people have more choices to digitize papers simply by capturing the text images of interest and put them through Optical Character Recognition (OCR) systems. However, the task is much more challenging when it comes to the issue of recognition of camera-based text images taken at remote distances. Those images, which often come up in small font and low resolution texts, results in poor readability in OCR. This paper presents a preliminary study on the human eye's perception of English document images, taken by mobile phone at various distances. In addition, a text enhancement technique using the combination of power-law transformation and morphological bottom hat filtering, following by morphological reconstruction and interpolation is also addressed. The results achieve an improved performance in OCR recognition rate.

Constrained Optimum Design of 3-DOF Microaccelerometers

Abstract: This paper presents a technique to optimize with constraints, the performance of three degrees of freedom silicon accelerometers. A flexure configuration has been proposed in order to meet the requirements of small cross-axial acceleration, high sensitivity, and linear sensitivity. The overall chip dimension is 1.5 × 1.5 × 0.5 mm3 (L × W × T) and the beam size is 950 × 80 × 10 μm3 (L × W × T). The purpose of this constrained optimization process is to achieve the highest sensitivity or resolution while imposing conditions on other parameters. It has been done based on the considerations of the junction depth, the doping concentration of the piezoresistor, the temperature, the signal to noise ratio, and the power consumption. Such an optimized accelerometer offers a much better performance compared to others.

An independent character recognizer for distantly acquired mobile phone text images

Abstract: In the field of character recognition, OCR is a success as many software have been commercialized, thanks to the exhaustive efforts of scientists. Nevertheless, there are existing cases in which these sophisticated systems fail, which has not been excavated that demands further investigation. This paper presents an attempt to build an effective algorithm in response to a cutting-edge topic in OCR, to determine the content of writing of type-written text document images in a special situation, where the source images are acquired by mobile phone camera, at a distant location. Through evaluating by ABBYY Finereader, it is observed that our system produced an improved gain in success rate.

Two-Step CNN Framework for Text Line Recognition in Camera-Captured Images

Abstract: In this paper, we introduce an “on the device” text line recognition framework that is designed for mobile or embedded systems. We consider per-character segmentation as a language-independent problem and individual character recognition as a language-dependent one. Thus, the proposed solution is based on two separate artificial neural networks (ANN) and dynamic programming instead of employing image processing methods for the segmentation step or end-to-end ANN. To satisfy the tight constraints on memory size imposed by embedded systems and to avoid overfitting, we employ ANNs with a small number of trainable parameters. The primary purpose of our framework is the recognition of low-quality images of identity documents with complex backgrounds and a variety of languages and fonts. We demonstrate that our solution shows high recognition accuracy on natural datasets even being trained on purely synthetic data. We use MIDV-500 and Census 1961 Project datasets for text line recognition. The proposed method considerably surpasses the algorithmic method implemented in Tesseract 3.05, the LSTM method (Tesseract 4.00), and unpublished method used in the ABBYY FineReader 15 system. Also, our framework is faster than other compared solutions. We show the language-independence of our segmenter with the experiment with Cyrillic, Armenian, and Chinese text lines.

On the Use of Alpha-Stable Distributions in Noise Modeling for PLC

Abstract: Power-line communication (PLC) deals with the challenge of transmitting data over a channel impaired by background and impulsive noises. In this paper, we provide evidence that the marginal distribution of the noise found in PLC systems exhibits some statistical properties that can be well captured by using $\alpha$ -stable distribution. Motivated by this result, we present a practical application of the $\alpha$ -stable model to synthesize noise in the power line. Furthermore, we also show that this family of distributions is a suitable alternative in order to model the PLC background noise that is commonly assumed as Gaussian. Finally, we study the performance of an orthogonal frequency-division multiplexing communication system under both kinds of background noise (i.e., Gaussian and $\alpha$ -stable). From these experiments, we conclude that if highly impulsive background traffic is mistakenly assumed as Gaussian in evaluation studies, the system performance is largely overestimated.

Design and implementation of site-specific rainfall-induced landslide early warning and monitoring system: a case study at Nam Dan landslide (Vietnam)

Abstract: This paper proposes and implements an early warning and monitoring system for rainfall-induced landslide (named as EWMRIL) with a case study at the Nam Dan landslide (northern Vietnam). The propose...

Open-Source Wireless Sensor System for Long-Term Monitoring of Slope Movement

Abstract: Power consumption is one of the major issues associated with deploying a wireless sensor to monitor natural environments in the real world. It is not practical to frequently replace the battery of a sensor that is located in a remote mountainous area. While considering to save valuable energy, the sensor must be able to detect a particular event in a timely manner and report data. To satisfy these requirements, in this paper, we present a wireless sensor system, termed SMARTCONE, which is designed to monitor the slope movement and minimize the standby power consumption, while remaining active to detect events. Multiple SMARTCONEs need to synchronously change their operation modes to collect the physical parameters and transmit raw vibration data simultaneously. SMARTCONE is consuming a power of only 0.05 mA at 3.6 V in standby mode, which is significantly less than that consumed by the previous design. As far as we know, SMARTCONE is an unprecedented instrument for the monitoring of slope movement; no other instruments can achieve such a low-energy consumption while being triggered by external vibration and managed remotely. These complicated issues are considered while designing the SMARTCONE, whose performance is then evaluated. In order to prevent redundant effort and to reduce the entry-level knowledge required by the users who wish to design a system, and to promote the use of this design for monitoring natural environments, we would like to open source the design of SMARTCONE for the public. It can be modified for use in other applications to satisfy their requirements without the need to build from scratch.

Measuring Energy Consumption of a Wireless Sensor Node During Transmission: panStamp

Abstract: In most of the real-life application scenarios, where Wireless Sensor Networks (WSNs) are deployed, the energy is a limited and valuable resource. Many energy-based strategies were proposed in the literature to reduce the power consumption of sensor nodes and thus enhance the network lifetime. When those algorithms are developed and tested with some well-known network simulators like NS3, OPNET, the energy estimation aspect is not a key criterion, because these simulators have their own specific energy models. By implementing these algorithms in a real sensor node, the energy consumption can be measured and compared to the results provided by other previously implemented approaches. This paper proposes a method for estimating the energy consumption of a low power wireless node panStamp NRG 2.0. The proposed model calculates the consumed energy of the system, at run time for different operation cycles. Besides, since the biggest amount of energy is consumed during the communication, this paper aims to characterize the power consumption in the transmission state. A proper metric of the energy consumed per byte transmission pattern of the packets is provided. The results can be used to derive not only the energy consumption for other protocols but also the energy needed to transmit one byte as well as the payload size are given.