• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.3
• /
• pp.894-907
• /
• 2022

The features extracted by convolutional neural networks are more descriptive of images than traditional features, and their convolutional layers are more suitable for retrieving images than are fully connected layers. The convolutional layer features will consume considerable time and memory if used directly to match an image. Therefore, this paper proposes a feature weighting and region integration method for convolutional layer features to form global feature vectors and subsequently use them for image matching. First, the 3D feature of the last convolutional layer is extracted, and the convolutional feature is subsequently weighted again to highlight the edge information and position information of the image. Next, we integrate several regional eigenvectors that are processed by sliding windows into a global eigenvector. Finally, the initial ranking of the retrieval is obtained by measuring the similarity of the query image and the test image using the cosine distance, and the final mean Average Precision (mAP) is obtained by using the extended query method for rearrangement. We conduct experiments using the Oxford5k and Paris6k datasets and their extended datasets, Paris106k and Oxford105k. These experimental results indicate that the global feature extracted by the new method can better describe an image.

• Journal of the Korean Graphic Arts Communication Society
• /
• v.22 no.2
• /
• pp.73-82
• /
• 2004

A thermal transfer method was developed novel method to form the phosphor screen for monochrom VPT. This method have advantages of simple process, clean environment, saving raw material and running-cost. But now applying phosphor screen for thermal transfer method has been formed three layers (phosphor layer, ITO layer and thermal adhesive layer) on the PET film as substrate. This is complex process, consumption of raw-material and require of high cost. Also ITO paste at present has been imported from Japan. To improve these problems, we have developed ITO paste as conductive paste by using ITO sol and binder resin (AA3003). Ito paste as developed in this study has both conductive and excellent thermal transfer abilities. Thus we could manufacture phosphor screen formed two layers (phosphor layer and ITO layer).

• Journal of IKEEE
• /
• v.25 no.3
• /
• pp.493-500
• /
• 2021

In this paper, we propose the development of deep learning structure to improve quality of polygonal containers. The deep learning structure consists of a convolution layer, a bottleneck layer, a fully connect layer, and a softmax layer. The convolution layer is a layer that obtains a feature image by performing a convolution 3x3 operation on the input image or the feature image of the previous layer with several feature filters. The bottleneck layer selects only the optimal features among the features on the feature image extracted through the convolution layer, reduces the channel to a convolution 1x1 ReLU, and performs a convolution 3x3 ReLU. The global average pooling operation performed after going through the bottleneck layer reduces the size of the feature image by selecting only the optimal features among the features of the feature image extracted through the convolution layer. The fully connect layer outputs the output data through 6 fully connect layers. The softmax layer multiplies and multiplies the value between the value of the input layer node and the target node to be calculated, and converts it into a value between 0 and 1 through an activation function. After the learning is completed, the recognition process classifies non-circular glass bottles by performing image acquisition using a camera, measuring position detection, and non-circular glass bottle classification using deep learning as in the learning process. In order to evaluate the performance of the deep learning structure to improve quality of polygonal containers, as a result of an experiment at an authorized testing institute, it was calculated to be at the same level as the world's highest level with 99% good/defective discrimination accuracy. Inspection time averaged 1.7 seconds, which was calculated within the operating time standards of production processes using non-circular machine vision systems. Therefore, the effectiveness of the performance of the deep learning structure to improve quality of polygonal containers proposed in this paper was proven.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2005.07a
• /
• pp.390-392
• /
• 2005

The digital radiation detectors are used clinically by diagnostic apparatus. However the digital radiation detector are some problem like high operating voltage, light blurring, low conversion efficiency, low fill factor, etc. Thus we propose a new radiation detector that the photoconductor layer and liquid crystal layer are coupled in sandwich structure. X-ray absorption in the photoconductor layer controls the state of the liquid crystal via creation of charge carrier and the light modulation of liquid crystal make image formation. The advantage of the new radiation detector is that high resolution image is acquired and the signal amplification is possible by external visible light source. In this study, we study the optical properties and electrical properties of the new radiation detector to irradiate X-ray. The Mercury Iodide($HgI_2$) was used by photoconductor material, and the aluminum is used by reflective layer. The thickness of Mercury Iodide is about $200{\mu}m$, the operating voltage of the liquid crystal is 1.5~5V. The electrical properties of Mercury Iodide was measured, and the transmission efficiency of liquid crystal was measured by modulation potential.

• The Journal of the Korea Contents Association
• /
• v.13 no.12
• /
• pp.38-47
• /
• 2013

An image coding based on MRC model, a kind of multi-layer image model, first segments a screen image into foreground, mask, and background layers, and then compresses each layer using a codec that is suitable to the layer. The mask layer defines the position of foreground regions such as textual and graphical contents. The colour signal of the foreground (background) region is saved in the foreground (background) layer. The mask layer which contains the segmentation result of foreground and background regions is of importance since its accuracy directly affects the overall coding performance of the codec. This paper proposes a new layer segmentation algorithm for the MRC based image coding. The proposed method extracts text pixels from the background using morphological top hat filtering. The application of white or black top hat transformation to local blocks is controlled by the information of relative brightness of text compared to the background. In the proposed method, the boundary information of text that is extracted from the edge map of the block is used for the robust decision on the relative brightness of text. Simulation results show that the proposed method is superior to the conventional methods.

• 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.

• International Journal of Advanced Culture Technology
• /
• v.6 no.3
• /
• pp.142-150
• /
• 2018

This paper describes the algorithm that lowers the dimension, maintains the object recognition and significantly reduces the eigenspace configuration time by combining the edge orientation histogram and principle component analysis. By using the detected object region as a recognition input image, in this paper the object recognition method combined with principle component analysis and the multi-layer network which is one of the intelligent classification was suggested and its performance was evaluated. As a pre-processing algorithm of input object image, this method computes the eigenspace through principle component analysis and expresses the training images with it as a fundamental vector. Each image takes the set of weights for the fundamental vector as a feature vector and it reduces the dimension of image at the same time, and then the object recognition is performed by inputting the multi-layer neural network.

• ETRI Journal
• /
• v.39 no.6
• /
• pp.841-850
• /
• 2017

We investigate neural network image reconstruction for magnetic particle imaging. The network performance strongly depends on the convolution effects of the spectrum input data. The larger convolution effect appearing at a relatively smaller nanoparticle size obstructs the network training. The trained single-layer network reveals the weighting matrix consisting of a basis vector in the form of Chebyshev polynomials of the second kind. The weighting matrix corresponds to an inverse system matrix, where an incoherency of basis vectors due to low convolution effects, as well as a nonlinear activation function, plays a key role in retrieving the matrix elements. Test images are well reconstructed through trained networks having an inverse kernel matrix. We also confirm that a multi-layer network with one hidden layer improves the performance. Based on the results, a neural network architecture overcoming the low incoherence of the inverse kernel through the classification property is expected to become a better tool for image reconstruction.

• Current Optics and Photonics
• /
• v.7 no.2
• /
• pp.200-206
• /
• 2023

Near-infrared (NIR) sensing technology using CMOS image sensors is used in many applications, including automobiles, biological inspection, surveillance, and mobile devices. An intuitive way to improve NIR sensitivity is to thicken the light absorption layer (silicon). However, thickened silicon lacks NIR sensitivity and has other disadvantages, such as diminished optical performance (e.g. crosstalk) and difficulty in processing. In this paper, a pixel structure for NIR sensing using a stacked CMOS image sensor is introduced. There are two photodetection layers, a conventional layer and a bottom photodiode, in the stacked CMOS image sensor. The bottom photodiode is used as the NIR absorption layer. Therefore, the suggested pixel structure does not change the thickness of the conventional photodiode. To verify the suggested pixel structure, sensitivity was simulated using an optical simulator. As a result, the sensitivity was improved by a maximum of 130% and 160% at wavelengths of 850 nm and 940 nm, respectively, with a pixel size of 1.2 ㎛. Therefore, the proposed pixel structure is useful for NIR sensing without thickening the silicon.

• Journal of Information Processing Systems
• /
• v.14 no.6
• /
• pp.1347-1360
• /
• 2018

Owing to limited energy in wireless devices power saving is very critical to prolong the lifetime of the networks. In this regard, we designed a cross-layer optimization mechanism based on power control in which source node broadcasts a Route Request Packet (RREQ) containing information such as node id, image size, end to end bit error rate (BER) and residual battery energy to its neighbor nodes to initiate a multimedia session. Each intermediate node appends its remaining battery energy, link gain, node id and average noise power to the RREQ packet. Upon receiving the RREQ packets, the sink node finds node disjoint paths and calculates the optimal power vectors for each disjoint path using cross layer optimization algorithm. Sink based cross-layer maximal minimal residual energy (MMRE) algorithm finds the number of image packets that can be sent on each path and sends the Route Reply Packet (RREP) to the source on each disjoint path which contains the information such as optimal power vector, remaining battery energy vector and number of packets that can be sent on the path by the source. Simulation results indicate that considerable energy saving can be accomplished with the proposed cross layer power control algorithm.