• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.7
• /
• pp.29-38
• /
• 2009

In this paper, a high performance pipelined computing design of parallel multiplier-temporal buffer-parallel accumulator is present for the convolution-based non-cascaded architecture aiming at the real time Discrete Wavelet Transform(DWT) processing. The convolved multiplication of DWT would be reduced upto 1/4 by utilizing the filter coefficients symmetry and the up/down sampling; and it could be dealt with 3-5 times faster computation by LUT-based DA multiplication of multiple filter coefficients parallelized for product terms with an image data. Further, the reutilization of computed product terms could be achieved by storing in the temporal buffer, which yields the saving of computation as well as dynamic power by 50%. The convolved product terms of image data and filter coefficients are realigned and stored in the temporal buffer for the accumulated addition. Then, the buffer management of parallel aligned storage is carried out for the high speed sequential retrieval of parallel accumulations. The convolved computation is pipelined with parallel multiplier-temporal buffer-parallel accumulation in which the parallelization of temporal buffer and accumulator is optimize, with respect to the performance of parallel DA multiplier, to improve the pipelining performance. The proposed architecture is back-end designed with 0.18um library, which verifies the 30fps throughput of SVGA(800$\times$600) images at 90MHz.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.3
• /
• pp.25-33
• /
• 2023

Noise generated during the acquisition and transmission of CT images acts as a factor that degrades image quality. Therefore, noise removal to solve this problem is an important preprocessing process in image processing. In this paper, we remove noise by using a deformable convolutional autoencoder (DeCAE) model in which deformable convolution operation is applied instead of the existing convolution operation in the convolutional autoencoder (CAE) model of deep learning. Here, the deformable convolution operation can extract features of an image in a more flexible area than the conventional convolution operation. The proposed DeCAE model has the same encoder-decoder structure as the existing CAE model, but the encoder is composed of deformable convolutional layers and the decoder is composed of conventional convolutional layers for efficient noise removal. To evaluate the performance of the DeCAE model proposed in this paper, experiments were conducted on CT images corrupted by various noises, that is, Gaussian noise, impulse noise, and Poisson noise. As a result of the performance experiment, the DeCAE model has more qualitative and quantitative measures than the traditional filters, that is, the Mean filter, Median filter, Bilateral filter and NL-means method, as well as the existing CAE models, that is, MAE (Mean Absolute Error), PSNR (Peak Signal-to-Noise Ratio) and SSIM. (Structural Similarity Index Measure) showed excellent results.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.3
• /
• pp.283-288
• /
• 2014

The use of robots is no longer limited to the field of industrial robots and is now expanding into the fields of service and medical robots. In this light, a trajectory generation method that can respond instantaneously to the external environment is strongly required. Toward this end, this study proposes a method that enables a robot to change its trajectory in real-time using a convolution operation. The proposed method generates a trajectory in real time and satisfies the physical limits of the robot system such as acceleration and velocity limit. Moreover, a new way to improve the previous method (11), which generates inefficient trajectories in some cases owing to the characteristics of the trapezoidal shape of trajectories, is proposed by introducing a triangle shape. The validity and effectiveness of the proposed method is shown through a numerical simulation and a comparison with the previous convolution method.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.4
• /
• pp.511-523
• /
• 2023

Research on the integration of unmanned aerial vehicles and deep learning for reinforced concrete damage detection is actively underway. Convolutional neural networks have a high impact on the performance of image classification, detection, and segmentation as backbones. The MobileNet, a pre-trained convolutional neural network, is efficient as a backbone for an unmanned aerial vehicle-based damage detection model because it can achieve sufficient accuracy with low computational complexity. Analyzing vanilla convolutional neural networks and MobileNet under various conditions, MobileNet was evaluated to have a verification accuracy 6.0~9.0% higher than vanilla convolutional neural networks with 15.9~22.9% lower computational complexity. MobileNetV2, MobileNetV3Large and MobileNetV3Small showed almost identical maximum verification accuracy, and the optimal conditions for MobileNet's reinforced concrete damage image feature extraction were analyzed to be the optimizer RMSprop, no dropout, and average pooling. The maximum validation accuracy of 75.49% for 7 types of damage detection based on MobilenetV2 derived in this study can be improved by image accumulation and continuous learning.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1238-1241
• /
• 2018

In this paper, MNIST classification CNN with zero skipping is implemented. Activation of CNN results in 30% to 40% zero. Since 0 does not affect the MAC operation, skipping 0 through a branch can improve performance. However, at the convolution layer, skipping over a branch causes a performance degradation. Accordingly, in the convolution layer, an operation is skipped by giving a NOP that does not affect the operation. Fully connected layer is skipped through the branch. We have seen performance improvements of about 1.5 times that of existing CNN.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.4
• /
• pp.161-166
• /
• 2018

In this paper, we propose a method for the random selection of pooling operations for the regularization and reduction of cross validation in convolutional neural networks. The pooling operation in convolutional neural networks is used to reduce the size of the feature map and for its shift invariant properties. In the existing pooling method, one pooling operation is applied in each pooling layer. Because this method fixes the convolution network, the network suffers from overfitting, which means that it excessively fits the models to the training samples. In addition, to find the best combination of pooling operations to maximize the performance, cross validation must be performed. To solve these problems, we introduce the probability concept into the pooling layers. The proposed method does not select one pooling operation in each pooling layer. Instead, we randomly select one pooling operation among multiple pooling operations in each pooling region during training, and for testing purposes, we use probabilistic weighting to produce the expected output. The proposed method can be seen as a technique in which many networks are approximately averaged using a different pooling operation in each pooling region. Therefore, this method avoids the overfitting problem, as well as reducing the amount of cross validation. The experimental results show that the proposed method can achieve better generalization performance and reduce the need for cross validation.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.567-572
• /
• 2019

In this paper, we propose an algorithm to perform convolution, pooling, and ReLU operations in CNN using binary image and binary kernel. It decomposes 256 gray-scale images into 8 bit planes and uses a binary kernel consisting of -1 and 1. The convolution operation of binary image and binary kernel is performed by addition and subtraction. Logically, it is a binary operation algorithm using the XNOR and comparator. ReLU and pooling operations are performed by using XNOR and OR logic operations, respectively. Through the experiments to verify the usefulness of the proposed algorithm, We confirm that the CNN operation can be performed by converting it to binary logic operation. It is an algorithm that can implement deep running even in a system with weak computing power. It can be applied to a variety of embedded systems such as smart phones, intelligent CCTV, IoT system, and autonomous car.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.4
• /
• pp.288-294
• /
• 2017

Recently, deep learning using convolutional neural network (CNN) has been extensively studied in image recognition. Convolution consists of addition and multiplication. Multiplication is computationally expensive in hardware implementation, relative to addition. It is also important factor limiting a chip design in an embedded deep learning system. In this paper, I propose a parallel-addition processing algorithm that converts grayscale images to the superposition of binary images and performs convolution only with addition. It is confirmed that the convolution can be performed by a parallel-addition method capable of reducing the processing time in experiment for verifying the availability of proposed algorithm.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.354-357
• /
• 2013

The filtering algorithm is used very frequently in the preprocessing stage of many image processing algorithms in computer vision processing. Because video signals are two-dimensional signals, computaional complexity is very high. To reduce the complexity, separable filters and the factorization theorem is applied to the filtering operation. As a result, it is shown that a significant reduction in computational complexity is achieved, although the experimental results could be slightly different depending on the condition of the image.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1250-1256
• /
• 2019

Although one can easily generate real-world 3D mesh data using a 3D printer or a depth camera, the generated data inevitably includes unnecessary noise. Therefore, mesh denoising is essential to obtain intact 3D mesh data. However, conventional mathematical denoising methods require preprocessing and often eliminate some important features of the 3D mesh. To address this problem, this paper proposes a deep learning based 3D mesh denoising method. Specifically, we propose a convolution-based autoencoder model consisting of an encoder and a decoder. The convolution operation applied to the mesh data performs denoising considering the relationship between each vertex constituting the mesh data and the surrounding vertices. When the convolution is completed, a sampling operation is performed to improve the learning speed. Experimental results show that the proposed autoencoder model produces faster and higher quality denoised data than the conventional methods.