• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.8
• /
• pp.136-144
• /
• 2016

In general, it is this variety of learning methods and teaching tools for embedded software development and deployment training. In this paper, I want to take advantage of the smart robot to learn creative problem-solving methods that are required in embedded software engineering education. It analyzes the effect of creative engineering education with the smart robot and presents for RSPM Engineering Embedded SW teaching methods to improve NCS education. Embedded SW engineering education in a more creative and smart robot, EV3 system was utilized to improve SW programming skills. In this paper, we utilize the EV3 system to the parish through the creative RSPM engineering courses through the survey and analysis of the impact level, interests and program skills and influence in embedded SW engineering education propose for successful embedded software programming skills potential.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.17 no.3
• /
• pp.9-15
• /
• 2021

In this paper, the Embedded MEMS system to the power apparatus used Edge Impulse machine learning tools and therefore an improved predictive system design is implemented. The proposed MEMS embedded system is developed based on nRF52840 system and the sensor with 3-Axis Digital Magnetometer, I2C interface and magnetic measurable range ±120 uT, BM1422AGMV which incorporates magneto impedance elements to detect magnetic field and the ARM M4 32-bit processor controller circuit in a small package. The MEMS embedded platform is consisted with Edge Impulse Machine Learning and system driver implementation between hardware and software drivers using SensorQ which is special queue including user application temporary sensor data. In this paper by experimenting, TensorFlow machine learning training output is applied to the power apparatus for analyzing the status such as "Normal, Warning, Hazard" and predicting the performance at level of 99.6% accuracy and 0.01 loss.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.10
• /
• pp.23-29
• /
• 2009

This paper presents the development of embedded network educational system. This is an educational equipment which enables user to have training over Network Configuration and Embedded network programming practice on Internet environment. The network education system is developed on embedded environment. based on using ethernet interface. On the development environment. PAX255 VLSI chip is used for the processor, the ADSv1.2 for debugging, uC/OS276 for RTOS. The system software was developed using C language. The ping program provided an educational environment for the student to compile and load it to run after doing practice of demonstration behavior. Afterwards programming procedure starts the step-by-step training just like the demonstration function. In other words, programming method how to design the procedure of ARP operation and ICMP operation is explained.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.9 no.2
• /
• pp.87-99
• /
• 2014

An AUTomotive Open System ARchitecture (AUTOSAR) is a de-facto standardized software platform, which developed for an automotive Electronic Control Unit (ECU) in global automotive industry. AUTOSAR improves the reusability and the scalability, thus the software development can be easier, faster and more reliable. However, it requires a lot of time and efforts to develop an AUTOSAR software due to the difficulties of understanding of massive AUTOSAR documentations and complicated usage of AUTOSAR design tools. AUTOSAR training is offered by AUTOSAR design tool venders but it is limited to introduction of their simplified concept and usages based on PC. Therefore the training is not enough for industrial developers or graduate students. In this paper we present an AUTOSAR starter kit which allows industrial engineers and graduate students to practice the detailed process of AUTOSAR software development easily and more conveniently. The kit is composed of a practical environment similar to actual automotive system and a textbook that explains how to design AUTOSAR software. And we demonstrated the validity of our methodology based on a case study.

• Journal of Korea Multimedia Society
• /
• v.20 no.5
• /
• pp.769-781
• /
• 2017

Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground-Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.1
• /
• pp.1-8
• /
• 2021

Detecting Out-of-Distribution (OOD) data is fundamentally required when Deep Neural Network (DNN) is applied to real-world AI such as autonomous driving. However, modern DNNs are quite vulnerable to the over-confidence problem even if the test data are far away from the trained data distribution. To solve the problem, this paper proposes a novel Adversarial-Mixup training method to let the DNN model be more robust by detecting OOD data effectively. Experimental results show that the proposed Adversarial-Mixup method improves the overall performance of OOD detection by 78% comparing with the State-of-the-Art methods. Furthermore, we show that the proposed method can alleviate the over-confidence problem by reducing the confidence score of OOD data than the previous methods, resulting in more reliable and robust DNNs.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.3
• /
• pp.490-493
• /
• 2021

This paper proposes an embedded system that detects mask and face recognition based on a microprocessor instead of Nvidia Jetson Board what is popular development kit. We use a class of efficient models called Mobilenets for mobile and embedded vision applications. MobileNets are based on a streamlined architechture that uses depthwise separable convolutions to build light weight deep neural networks. The device used a Maix development board with CNN hardware acceleration function, and the training model used MobileNet_V2 based SSD(Single Shot Multibox Detector) optimized for mobile devices. To make training model, 7553 face data from Kaggle are used. As a result of test dataset, the AUC (Area Under The Curve) value is as high as 0.98.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.5
• /
• pp.289-296
• /
• 2022

In this paper, we developed a deep learning-based recyclable object detection model. The model is developed based on YOLOv5 that is a one-stage detector. The deep learning model detects and classifies the recyclable object into 7 categories: paper, carton, can, glass, pet, plastic, and vinyl. We propose two methods for recyclable object detection models to solve problems during training. Bounding Box CutMix solved the no-objects training images problem of Mosaic, a data augmentation used in YOLOv5. Standardized Distance-based IoU replaced DIoU using a normalization factor that is not affected by the center point distance of the bounding boxes. The recyclable object detection model showed a final mAP performance of 0.91978 with Bounding Box CutMix and 0.91149 with Standardized Distance-based IoU.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.4
• /
• pp.209-215
• /
• 2022

This paper proposes a novel approach for training an unsupervised depth estimation algorithm. The objective of unsupervised depth estimation is to estimate pixel-wise distances from camera without external supervision. While most previous works focus on model architectures, loss functions, and masking methods for considering dynamic objects, this paper focuses on the training framework to effectively use depth cue. The main loss function of unsupervised depth estimation algorithms is known as the photometric error. In this paper, we claim that direct depth cue is more effective than the photometric error. To obtain the direct depth cue, we adopt the technique of knowledge distillation which is a teacher-student learning framework. We train a teacher network based on a previous unsupervised method, and its depth predictions are utilized as pseudo labels. The pseudo labels are employed to train a student network. In experiments, our proposed algorithm shows a comparable performance with the state-of-the-art algorithm, and we demonstrate that our teacher-student framework is effective in the problem of unsupervised depth estimation.

• The Journal of Korea Robotics Society
• /
• v.17 no.2
• /
• pp.133-141
• /
• 2022

This paper introduces model compression algorithms which make a deep neural network smaller and faster for embedded systems. The model compression algorithms can be largely categorized into pruning, quantization and knowledge distillation. In this study, gradual pruning, quantization aware training, and knowledge distillation which learns the activation boundary in the hidden layer of the teacher neural network are integrated. As a large deep neural network is compressed and accelerated by these algorithms, embedded computing boards can run the deep neural network much faster with less memory usage while preserving the reasonable accuracy. To evaluate the performance of the compressed neural networks, we evaluate the size, latency and accuracy of the deep neural network, DenseNet201, for image classification with CIFAR-10 dataset on the NVIDIA Jetson Xavier.