• Journal of the Korea Society of Computer and Information
• /
• v.29 no.7
• /
• pp.41-51
• /
• 2024

In this paper, we explore the enhancement of target detection accuracy in the guided weapon using deep learning object detection on infrared (IR) images. Due to the characteristics of IR images being influenced by factors such as time and temperature, it's crucial to ensure a consistent representation of object features in various environments when training the model. A simple way to address this is by emphasizing the features of target objects and reducing noise within the infrared images through appropriate pre-processing techniques. However, in previous studies, there has not been sufficient discussion on pre-processing methods in learning deep learning models based on infrared images. In this paper, we aim to investigate the impact of image pre-processing techniques on infrared image-based training for object detection. To achieve this, we analyze the pre-processing results on infrared images that utilized global or local information from the video and the image. In addition, in order to confirm the impact of images converted by each pre-processing technique on object detector training, we learn the YOLOX target detector for images processed by various pre-processing methods and analyze them. In particular, the results of the experiments using the CLAHE (Contrast Limited Adaptive Histogram Equalization) shows the highest detection accuracy with a mean average precision (mAP) of 81.9%.

• Journal of Korea Multimedia Society
• /
• v.8 no.1
• /
• pp.46-55
• /
• 2005

In this paper, we realize the panoramic infrared warning system to detect the small threaten object and propose the infrared image enhancement method to improve the warning ability of this system. This system composes of the sense head unit, the signal processing unit, and so on. In the proposed system, the sense head unit acquires the panoramic IR image with 360 degree field of view(FOV) by rotating the thermal sensor. The signal processing unit divides panoramic image into four sub-images with 90 degree FOV and computes the adaptive plateau value by using statistical characteristics of each subimage. Then the histogram equalization is performed for each subimage by using the adaptive plateau value. We realize the signal Processing unit by using the DSP and FPGA to perform the proposed method in real time. Experimental results show that the proposed method has better discrimination and lower false alarm rate than the conventional methods in this warning system.

• Journal of Digital Contents Society
• /
• v.16 no.6
• /
• pp.951-959
• /
• 2015

Scene text extraction is important because it offers some important information on different image based applications pouring in current smart generation. Edge-Enhanced MSER(Maximally Stable Extremal Regions) which enhances the boundaries using the canny operator after extracting the basic MSER shows excellent performance in terms of text extraction. But according to setting the threshold of the canny operator, the result images using Edge-Enhanced MSER are different, so there needs a method figuring out the threshold. In this paper, we propose a AEMSER(Adaptive Edge-enhanced MSER) that applies the method extracting the boundary using the middle value of histogram to Edge-Enhanced MSER to get the canny operator's threshold. The proposed method can acquire better result images than the existing methods because it extracts the area only for the obvious boundaries.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.149-160
• /
• 1997

This paper is concerned with the problem of speaker-adaptive speech synthes is method using a mapped codebook designed by fuzzy mapping on FLVQ (Fuzzy Learning Vector Quantization). The FLVQ is used to design both input and reference speaker's codebook. This algorithm is incorporated fuzzy membership function into the LVQ(learning vector quantization) networks. Unlike the LVQ algorithm, this algorithm minimizes the network output errors which are the differences of clas s membership target and actual membership values, and results to minimize the distances between training patterns and competing neurons. Speaker Adaptation in speech synthesis is performed as follow;input speaker's codebook is mapped a reference speaker's codebook in fuzzy concepts. The Fuzzy VQ mapping replaces a codevector preserving its fuzzy membership function. The codevector correspondence histogram is obtained by accumulating the vector correspondence along the DTW optimal path. We use the Fuzzy VQ mapping to design a mapped codebook. The mapped codebook is defined as a linear combination of reference speaker's vectors using each fuzzy histogram as a weighting function with membership values. In adaptive-speech synthesis stage, input speech is fuzzy vector-quantized by the mapped codcbook, and then FCM arithmetic is used to synthesize speech adapted to input speaker. The speaker adaption experiments are carried out using speech of males in their thirties as input speaker's speech, and a female in her twenties as reference speaker's speech. Speeches used in experiments are sentences /anyoung hasim nika/ and /good morning/. As a results of experiments, we obtained a synthesized speech adapted to input speaker.

• Journal of KIISE:Software and Applications
• /
• v.31 no.3
• /
• pp.276-292
• /
• 2004

In this paper. we present methods that extract lung regions from chest EBT(electron beam tomography) images then segment the extracted lung region into lung lobes. We use histogram based thresholding and mathematical morphology for extracting lung regions. For detecting pulmonary fissures, we use edge detector and knowledge-based search method. We suggest this edge detector, which uses adaptive filter scale, to work very well for real edge and insensitive for edge by noise. Our experiments showed about 95% accuracy or higher in extracting lung regions and about 5 pixel distance error in detecting pulmonary fissures.

• Journal of Korea Multimedia Society
• /
• v.9 no.11
• /
• pp.1455-1464
• /
• 2006

In this paper, we propose an adaptive image enhancement algorithm. The proposed algorithm is classified with the MIE technique for intensity enhancement of input image and MSE techniques for saturation enhancement. The MIE technique is proposed to control the gamut mapping problem and a sudden change in image-brightness while Luminance signal is processing, The MSE techniques are proposed to control de-saturation or over-saturation while chrominance signal is processing. The proposed algorithm is focused on processing preference color for human vision in order to generate better image quality than the algorithms focused on processing uniformly to whole images, This algorithm can be applied to a monitor, TV and other display devices for high quality image.

• Journal of Information Processing Systems
• /
• v.9 no.3
• /
• pp.405-424
• /
• 2013

This research paper proposes a secured, robust approach of information security using steganography. It presents two component based LSB (Least Significant Bit) steganography methods for embedding secret data in the least significant bits of blue components and partial green components of random pixel locations in the edges of images. An adaptive LSB based steganography is proposed for embedding data based on the data available in MSB's (Most Significant Bits) of red, green, and blue components of randomly selected pixels across smooth areas. A hybrid feature detection filter is also proposed that performs better to predict edge areas even in noisy conditions. AES (Advanced Encryption Standard) and random pixel embedding is incorporated to provide two-tier security. The experimental results of the proposed approach are better in terms of PSNR and capacity. The comparison analysis of output results with other existing techniques is giving the proposed approach an edge over others. It has been thoroughly tested for various steganalysis attacks like visual analysis, histogram analysis, chi-square, and RS analysis and could sustain all these attacks very well.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.160-171
• /
• 1994

The wide dynamic range and severely attenuated contrast in mediastinal area appearing in typical chest radiographs have often caused difficulties in effective visualization and diagnosis of lung diseases. This paper proposes a new adaptive image enhancement technique which potentially solves this problem and there by improves observer performance through image processing. In the proposed method image processing is applied to the chest radiograph with different processing parameters for the lung field and mediastinum adaptively since there are much differences in anatomical and imaging properties between these two regions. To achieve this the chest radiograph is divided into the lung and mediastinum by gray level thresholding using the cumulative histogram and the dynamic range compression and local contrast enhancement are carried out selectively in the mediastinal region. Thereafter a gray scale transformation is performed considering the JND(just noticeable difference) characteristic for effective image displa. The processed images showed apparenty improved contrast in mediastinum and maintained moderate brightness in the lung field. No artifact could be observed. In the visibility evaluation experiment with 5 radiologists the processed images with better visibility was observed for the 5 important anatomical structures in the thorax.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.1
• /
• pp.107-117
• /
• 2016

STD method using a mean and standard deviation is widely used in various inspection systems. The result of detection using the STD method is very dependent on the threshold value. This paper proposes an adaptive defect detection algorithm to with a precise detection of an ultimate defect. The proposed method is determined threshold value adaptively using a skewness that indicates a similarity of intensity and normal distribution of image. In the experiment, we used a various TFT-LCD images for a quantitative evaluation of defect detection performance evaluation result to prove the performance of the proposed algorithm.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2129-2147
• /
• 2019

This paper introduces an adaptive face recognition method based on a Novel Incremental Kernel Nonparametric Discriminant Analysis (IKNDA) that is able to learn through time. More precisely, the IKNDA has the advantage of incrementally reducing data dimension, in a discriminative manner, as new samples are added asynchronously. Thus, it handles dynamic and large data in a better way. In order to perform face recognition effectively, we combine the Gabor features and the ordinal measures to extract the facial features that are coded across local parts, as visual primitives. The variegated ordinal measures are extraught from Gabor filtering responses. Then, the histogram of these primitives, across a variety of facial zones, is intermingled to procure a feature vector. This latter's dimension is slimmed down using PCA. Finally, the latter is treated as a facial vector input for the advanced IKNDA. A comparative evaluation of the IKNDA is performed for face recognition, besides, for other classification endeavors, in a decontextualized evaluation schemes. In such a scheme, we compare the IKNDA model to some relevant state-of-the-art incremental and batch discriminant models. Experimental results show that the IKNDA outperforms these discriminant models and is better tool to improve face recognition performance.