• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2390-2406
• /
• 2022

Scene text recognition has important application value and attracted the interest of plenty of researchers. At present, many methods have achieved good results, but most of the existing approaches attempt to improve the performance of scene text recognition from the image level. They have a good effect on reading regular scene texts. However, there are still many obstacles to recognizing text on low-quality images such as curved, occlusion, and blur. This exacerbates the difficulty of feature extraction because the image quality is uneven. In addition, the results of model testing are highly dependent on training data, so there is still room for improvement in scene text recognition methods. In this work, we present a natural scene text recognizer to improve the recognition performance from the feature level, which contains feature representation and feature enhancement. In terms of feature representation, we propose an efficient feature extractor combined with Representative Batch Normalization and ResNet. It reduces the dependence of the model on training data and improves the feature representation ability of different instances. In terms of feature enhancement, we use a feature enhancement network to expand the receptive field of feature maps, so that feature maps contain rich feature information. Enhanced feature representation capability helps to improve the recognition performance of the model. We conducted experiments on 7 benchmarks, which shows that this method is highly competitive in recognizing both regular and irregular texts. The method achieved top1 recognition accuracy on four benchmarks of IC03, IC13, IC15, and SVTP.

• The KIPS Transactions:PartB
• /
• v.12B no.7 s.103
• /
• pp.721-728
• /
• 2005

This paper suggests effective object detection system for moving objects with specified color and motion information. The proposed detection system includes the object extraction and definition process which uses MCN(Multiple Color Normalization) and MCWUPC(Moving Color Weighted Unmatched Pixel Count) computation to decide the existence of moving object and object segmentation technique using signature information is used to exactly extract the objects with high probability. Finally, real time detection system is implemented to verify the effectiveness of the technique and experiments show that the success rate of object tracking is more than $89\%$ of total 120 image frames.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.51-62
• /
• 2021

Plant disease is one of the issues that can create losses in the production and economy of the agricultural sector. Early detection of this disease for finding solutions and treatments is still a challenge in the sustainable agriculture field. Currently, image processing techniques and machine learning methods have been applied to detect plant diseases successfully. However, the effectiveness of these methods still needs to be improved, especially in multiclass plant diseases classification. In this paper, a convolutional neural network with a batch normalization-based deep learning approach for classifying plant diseases is used to develop an automatic diagnostic assistance system for leaf diseases. The significance of using deep learning technology is to make the system be end-to-end, automatic, accurate, less expensive, and more convenient to detect plant diseases from their leaves. For evaluating the proposed model, an experiment is conducted on a public dataset contains 20654 images with 15 plant diseases. The experimental validation results on 20% of the dataset showed that the model is able to classify the 15 plant diseases labels with 96.4% testing accuracy and 0.168 testing loss. These results confirmed the applicability and effectiveness of the proposed model for the plant disease detection task.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.2
• /
• pp.423-430
• /
• 2013

Recently, the personal identification technologies using vein pattern of back of the hand, palm, and finger have been developed actively because it has the advantage that the vein blood vessel in the body is impossible to damage, make a replication and forge. However, it is difficult to extract clearly the vein region from captured vein images through common image prcessing based region segmentation method, because of the light scattering and non-uniform internal tissue by skin layer and inside layer skeleton, etc. Especially, it takes a long time for processing time and makes a discontinuity of blood vessel just in a image because it has non-uniform illumination due to use a locally different adaptive threshold for the binarization of acquired finger-vein image. To solve this problem, we propose illumination normalization based fast method for extracting the finger-vein region. The proposed method has advantages compared to the previous methods as follows. Firstly, for remove a non-uniform illumination of the captured vein image, we obtain a illumination component of the captured vein image by using a low-pass filter. Secondly, by extracting the finger-vein path using one time binarization of a single threshold selection, we were able to reduce the processing time. Through experimental results, we confirmed that the accuracy of extracting the finger-vein region was increased and the processing time was shortened than prior methods.

• Journal of the Optical Society of Korea
• /
• v.18 no.4
• /
• pp.317-329
• /
• 2014

In vision measurement systems based on structured light, the key point of detection precision is to determine accurately the central position of the projected laser line in the image. The purpose of this research is to extract laser line centers based on a decision function generated to distinguish the real centers from candidate points with a high recognition rate. First, preprocessing of an image adopting a difference image method is conducted to realize image segmentation of the laser line. Second, the feature points in an integral pixel level are selected as the initiating light line centers by the eigenvalues of the Hessian matrix. Third, according to the light intensity distribution of a laser line obeying a Gaussian distribution in transverse section and a constant distribution in longitudinal section, a normalized model of Hessian matrix eigenvalues for the candidate centers of the laser line is presented to balance reasonably the two eigenvalues that indicate the variation tendencies of the second-order partial derivatives of the Gaussian function and constant function, respectively. The proposed model integrates a Gaussian recognition function and a sinusoidal recognition function. The Gaussian recognition function estimates the characteristic that one eigenvalue approaches zero, and enhances the sensitivity of the decision function to that characteristic, which corresponds to the longitudinal direction of the laser line. The sinusoidal recognition function evaluates the feature that the other eigenvalue is negative with a large absolute value, making the decision function more sensitive to that feature, which is related to the transverse direction of the laser line. In the proposed model the decision function is weighted for higher values to the real centers synthetically, considering the properties in the longitudinal and transverse directions of the laser line. Moreover, this method provides a decision value from 0 to 1 for arbitrary candidate centers, which yields a normalized measure for different laser lines in different images. The normalized results of pixels close to 1 are determined to be the real centers by progressive scanning of the image columns. Finally, the zero point of a second-order Taylor expansion in the eigenvector's direction is employed to refine further the extraction results of the central points at the subpixel level. The experimental results show that the method based on this normalization model accurately extracts the coordinates of laser line centers and obtains a higher recognition rate in two group experiments.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_2
• /
• pp.917-927
• /
• 2020

The Ministry of Land, Infrastructure and Transport of Korea is developing a CAS 500-1/2 satellite capable of photographing a GSD 0.5 m level image, and is developing a technology to utilize this. Therefore, this study attempted to develop a geospatial feature extraction technique aimed at automation as a technique for utilizing CAS 500-1/2 satellite images. KOMPSAT-3A satellite images that are expected to be most similar to CAS 500-1/2 were used for research and the possibility of automation of geospatial feature extraction was analyzed through relative radiometric normalization. For this purpose, the parameters and thresholds were applied equally to the reference images and relative radiometric normalized images, and the geospatial feature were extracted. The qualitative analysis was conducted on whether the extracted geospatial feature is extracted in a similar form from the reference image and relative radiometric normalized image. It was also intended to analyze the possibility of automation of geospatial feature extraction by quantitative analysis of whether the classification accuracy satisfies the target accuracy of 90% or more set in this study. As a result, it was confirmed that shape of geospatial feature extracted from reference image and relative radiometric normalized image were similar, and the classification accuracy analysis results showed that both satisfies the target accuracy of 90% or more. Therefore, it is believed that automation will be possible when extracting spatial objects through relative radiometric normalization.

• Proceedings of the IEEK Conference
• /
• 2005.11a
• /
• pp.367-370
• /
• 2005

This paper proposes a novel method for content-based watermarking based on feature points of an image. At each feature point, watermark is embedded after affine normalization according to the local characteristic scale and orientation. The characteristic scale is the scale at which the normalized scale-space representation of an image attains a maximum value, and the characteristic orientation is the angle of the principal axis of an image. By binding watermarking with the local characteristics of an image, resilience against affine transformations can be obtained. Experimental results show that the proposed method is robust against various image processing steps including affine transformations, cropping, filtering, and JPEG compression.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.219-224
• /
• 2017

This paper is to introduce an application of face recognition algorithm in parallel. We have experiments of 25 images with different motions and simulated the image recognitions; grouping of the image vectors, image normalization, calculating average image vectors, etc. We also discuss an analysis of the related eigen-image vectors and a parallel algorithm. To develop the parallel algorithm, we propose a new type of initial matrices for eigenvalue problem. If A is a symmetric matrix, initial matrices for eigen value problem are investigated: the "optimal" one, which minimize ${\parallel}C-A{\parallel}_F$ and the "super optimal", which minimize ${\parallel}I-C^{-1}A{\parallel}_F$. In this paper, we present a general new approach to the design of an initial matrices to solving eigenvalue problem based on the new optimal investigating C with preserving the characteristic of the given matrix A. Fast all resulting can be inverted via fast transform algorithms with O(N log N) operations.

• Proceedings of the KIEE Conference
• /
• 1995.11a
• /
• pp.225-227
• /
• 1995

An object detection algorithm using a modified IDM(Image Differential Method) is proposed for detecting an object in a level crossing area. The conventional object detection method using LASER light has the deadzone that it cannot detect small objects, while the object detection method using image data in a level crossing area can detect such small objects. But the image data in a level crossing area can be changeable easily because the data is outdoor and sensitive to such surrounding environments as the change of the sun beam, the shadow of cars, and so on. So we resolve these problems by adding the normalization and the process for shadow of the image data in a level crossing area to the basic IDM(Image Differential Method).

• Journal of the Korean Society of Clothing and Textiles
• /
• v.36 no.7
• /
• pp.663-676
• /
• 2012

This article is to understand the implications and ideological meaning of female normative beauty reproduced by the idealizing phenomena of the hyper-real body as a process of the normalization of the body projected in fashion magazines with a focus on the body created by the increased influence of mass media in consumer capitalism. This study conducts a literature research and semiotic analysis as the method of investigation and focuses on the body images of the beauty articles in Vogue Korea. The idealizing phenomena of the hyper-real body in fashion magazines emphasizes that the body is an exchangeable substance that can be disassembled to adjust to accord with the standards and norms of society, that the ability of individuals to manage their body is enhanced by a rise in social class, and concludes that the superficial alteration of the body image is related to the standard of a moral tendency where a young and slender figure is considered to be a well managed body image.