• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.5
• /
• pp.7-14
• /
• 2021

Tomato crops are highly affected by tomato diseases, and if not prevented, a disease can cause severe losses for the agricultural economy. Therefore, there is a need for a system that quickly and accurately diagnoses various tomato diseases. In this paper, we propose a system that classifies nine diseases as well as healthy tomato plants by applying various pretrained deep learning-based CNN models trained on an ImageNet dataset. The tomato leaf image dataset obtained from PlantVillage is provided as input to ResNet, Xception, and DenseNet, which have deep learning-based CNN architectures. The proposed models were constructed by adding a top-level classifier to the basic CNN model, and they were trained by applying a 5-fold cross-validation strategy. All three of the proposed models were trained in two stages: transfer learning (which freezes the layers of the basic CNN model and then trains only the top-level classifiers), and fine-tuned learning (which sets the learning rate to a very small number and trains after unfreezing basic CNN layers). SGD, RMSprop, and Adam were applied as optimization algorithms. The experimental results show that the DenseNet CNN model to which the RMSprop algorithm was applied output the best results, with 98.63% accuracy.

• Journal of Broadcast Engineering
• /
• v.26 no.2
• /
• pp.125-131
• /
• 2021

In recent years, Convolutional Neural Networks (CNNs) have achieved outstanding performance in the fields of computer vision such as image classification, object detection, visual quality enhancement, etc. However, as huge amount of computation and memory are required in CNN models, there is a limitation in the application of CNN to low-power environments such as mobile or IoT devices. Therefore, the need for neural network compression to reduce the model size while keeping the task performance as much as possible has been emerging. In this paper, we propose a method to compress CNN models by combining matrix decomposition methods of LR (Low-Rank) approximation and CP (Canonical Polyadic) decomposition. Unlike conventional methods that apply one matrix decomposition method to CNN models, we selectively apply two decomposition methods depending on the layer types of CNN to enhance the compression performance. To evaluate the performance of the proposed method, we use the models for image classification such as VGG-16, RestNet50 and MobileNetV2 models. The experimental results show that the proposed method gives improved classification performance at the same range of 1.5 to 12.1 times compression ratio than the existing method that applies only the LR approximation.

• Journal of IKEEE
• /
• v.22 no.1
• /
• pp.14-20
• /
• 2018

The purpose of training a convolutional neural network (CNN) is to obtain weight factors that give high classification accuracies. The initial values of hyper-parameters affect the training results, and it is important to train a CNN with a suitable hyper-parameter set of a learning rate, a batch size, the initialization of weight factors, and an optimizer. We investigate the effects of a single hyper-parameter while others are fixed in order to obtain a hyper-parameter set that gives higher classification accuracies and requires shorter training time using a proposed VGG-like CNN for training since the VGG is widely used. The CNN is trained for four datasets of CIFAR10, CIFAR100, GTSRB and DSDL-DB. The effects of the normalization and the data transformation for datasets are also investigated, and a training scheme using merged datasets is proposed.

• Journal of Korea Multimedia Society
• /
• v.23 no.6
• /
• pp.756-764
• /
• 2020

A typical algorithm used for image analysis is the Convolutional Neural Network(CNN). R-CNN, Fast R-CNN, Faster R-CNN, etc. have been studied to improve the performance of the CNN, but they essentially require large amounts of data and high algorithmic complexity., making them inappropriate for small and medium-sized services. Therefore, in this paper, the image label prediction algorithm based on CNN with collaborative layer with low complexity, high accuracy, and small amount of data was proposed. The proposed algorithm was designed to replace the part of the neural network that is performed to predict the final label in the existing deep learning algorithm by implementing collaborative filtering as a layer. It is expected that the proposed algorithm can contribute greatly to small and medium-sized content services that is unsuitable to apply the existing deep learning algorithm with high complexity and high server cost.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.509-513
• /
• 2003

In this paper, we configured secure communication circuit with PSpice through Embedding Drive Synchronization using SC-CNN, SC-CNN provide us a good method to separate interconnected state variables of a system respectively and to make it possible to change current component to voltage component in the state variables.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.505-508
• /
• 2003

In this paper, we introduce embedding drive synchronigation method using SC-CNN. SC-CNN provide us a good method to separate interconnected state variables of a system respectively and to make it possible to change current component to voltage component in the state variables. Using this advantage, We suggest another synchronization method called Embedding Drive Synchronization. Proposed Embedding Drive Synchronization change general drive synchronization method a little bit. we put transmitted signal between into the input of the state variables of the received part, not entirely replacing it. and we showed it proved.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.151-154
• /
• 2018

본 논문에서는 CNN Seq2Seq 구조를 이용해 한국어 대화 시스템을 개발하였다. 기존 Seq2Seq는 RNN 혹은 그 변형 네트워크에 데이터를 입력하고, 입력이 완료된 후의 은닉 층의 embedding에 기반해 출력열을 생성한다. 우리는 CNN Seq2Seq로 입력된 발화에 대해 출력 발화를 생성하는 대화 모델을 학습하였고, 그 성능을 측정하였다. CNN에 대해서는 약 12만 발화 쌍을 이용하여 학습하고 1만 발화 쌍으로 실험하였다. 평가 결과 제안 모델이 기존의 RNN 기반 모델에 비해 우수한 결과를 보였다.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.243-246
• /
• 2018

최근 사용자 발화를 이해하고 그에 맞는 피드백을 생성할 수 있는 대화 시스템의 중요성이 증가하고 있다. 따라서 사용자 의도를 파악하기 위한 화행 분석은 대화 시스템의 필수적인 요소이다. 최근 많이 연구되는 심층 학습 기법은 모델이 데이터로부터 자질들을 스스로 추출한다는 장점이 있다. 발화 자체의 연속성과 화자간 상호 작용을 포착하기 위하여 CNN에 RNN을 결합한 CNN-RNN을 제안한다. 본 논문에서 제안한 계층 구조 어텐션 매커니즘 기반 CNN-RNN을 효과적으로 적용한 결과 워드 임베딩을 추가한 조건에서 가장 높은 성능인 91.72% 정확도를 얻었다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.11a
• /
• pp.74-76
• /
• 2017

In this paper, we introduce the CNN-based in-loop technology for HEVC, and analyze the performance of these algorithms through comparative experiments. The current in-loop filters in HEVC are composed of a deblocking filter that removes noise and a sample adaptive offset filter that compensates for signal offsets. A couple of CNN-based filters replacing the roles of these two algorithms are selected and compared.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.1017-1020
• /
• 2000

In this paper, a CNN using 1-dimensional chaos circuits with controllable nonlinear functions is proposed. The proposed CNN consists of $\p{times}q$ chaos circuits which are called cell circuits. The nonlinear functions of the cell circuits can be controlled by employing fuzzy scheme. Thanks to the controllability of the nonlinear functions, the proposed circuit can adjust transition behavior of the CNN electronically. Furthermore, the chaotic behavior of the cell circuit which is a portion of the proposed CNN is simple since the cell circuit is a 1-dimensional chaos circuit. To confirm the validity of the circuit design, SPICE simulations were performed concerning the proposed CNN.