• Korean Journal of Agricultural Science
• /
• v.45 no.4
• /
• pp.823-837
• /
• 2018

The objective of this study was to evaluate the firmness and the sweetness index (SI) of tomatoes with a hyperspectral imaging (HSI) technique within the wavelength range of 1000 - 1550 nm. The hyperspectral images of 95 tomatoes were acquired with a push-broom hyperspectral reflectance imaging system, from which the mean spectra of each tomato were extracted from the regions of interest. The reference firmness and sweetness index of the same sample was measured and calibrated with their corresponding spectral data by partial least squares (PLS) regression with different preprocessing methods. The calibration model developed by PLS regression based on the Savitzky-Golay second-derivative preprocessed spectra resulted in a better performance for both the firmness and the SI of the tomatoes compared to models developed by other preprocessing methods. The correlation coefficients ($R_{pred}$) were 0.82, and 0.74 with a standard error of prediction of 0.86 N, and 0.63, respectively. Then, the feature wavelengths were identified using a model-based variable selection method, i.e., variable importance in projection, from the PLS regression analyses. Finally, chemical images were derived by applying the respective regression coefficients on the spectral image in a pixel-wise manner. The resulting chemical images provided detailed information on the firmness and the SI of the tomatoes. The results show that the proposed HSI technique has potential for rapid and non-destructive evaluation of firmness and the sweetness index of tomatoes.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.11
• /
• pp.1468-1473
• /
• 2018

The importance of communication and data processing is increasing with the advance of the Fourth Industrial Revolution. Hence, the importance of video and data processing technologies, which directly influence the accuracy and reliability of equipment, is also increasing. In this research report we propose an algorithm for calculating the final output by estimating the standard deviation and estimate required for removing AWGN while adapting to changes in the frequency factors of video. This algorithm calculates the final output by checking an estimated value against the effective pixel range, which is obtained from the standard deviation of mask factors. Subsequently, the weighted value is computed, taking into account the filter output. To evaluate the functionality of this algorithm, it is compared with the most-commonly used present method through simulation. The simulation results show that the important features of the image are preserved and efficient noise cancellation performance is demonstrated.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.3
• /
• pp.254-260
• /
• 2019

Recently, the importance of signal processing has become gradually significant, as the frequency of video media increases in various fields. However, numerous kinds of noise generated in the transmission and reception processes can possibly affect the signal information, and the noise removal is for that reason essential as a preprocessing step. In this paper, we propose an algorithm to remove the mixed noise which is composed of impulse noise and AWGN. This algorithm is used for image restoration by noise judgment for efficient noise removal in a complex noise environment, and the noise is removed by considering spatial characteristics and pixel variations. Simulation results show that unlike existing methods, the algorithm has excellent noise cancellation characteristics by minimizing both noise effects and consequently eliminating the mixed noise; for objective judgment, we compared and analyzed the data using PSNR and profile.

• Korean Journal of Optics and Photonics
• /
• v.33 no.6
• /
• pp.303-309
• /
• 2022

Since the pixel size of the image sensor used in optical systems is gradually decreasing, the resolution specification of the optical system should be inevitably higher. If aberration change according to the eccentricity of a specific lens group occurs, only the amount of eccentricity of a specific lens group may be calculated with the traditional resolution adjustment method so that the aberration of the optical system is minimized to a certain extent. As a result, it is possible to increase the resolution of the optical system and to respond to a sensor with a large number of pixels. However, in the traditional method, there should be no change in specific aberration due to the eccentricity of a specific lens group. In this paper, we propose a new method to eliminate such a limitation of the traditional method in a camera optical system with a floating system, which is to choose and control the arbitrary two lens groups to easily minimize the eccentricity of the optical system in order to obtain an optical system with high resolution.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.9
• /
• pp.99-112
• /
• 2019

Cracks on pavement surfaces are critical signs and symptoms of the degradation of pavement structures. Image-based pavement crack detection is a challenging problem due to the intensity inhomogeneity, topology complexity, low contrast, and noisy texture background. In this paper, we address the problem of pavement crack detection and segmentation at pixel-level based on a Deep Neural Network (DNN) using gray-scale images. We propose a novel DNN architecture which contains a modified U-net network and a high-level features network. An important contribution of this work is the combination of these networks afforded through the fusion layer. To the best of our knowledge, this is the first paper introducing this combination for pavement crack segmentation and detection problem. The system performance of crack detection and segmentation is enhanced dramatically by using our novel architecture. We thoroughly implement and evaluate our proposed system on two open data sets: the Crack Forest Dataset (CFD) and the AigleRN dataset. Experimental results demonstrate that our system outperforms eight state-of-the-art methods on the same data sets.

• Journal of radiological science and technology
• /
• v.45 no.2
• /
• pp.151-158
• /
• 2022

The Cardiac Gated Blood Pool (GBP) scintigram, a nuclear medicine imaging, calculates the left ventricular Ejection Fraction (EF) by segmenting the left ventricle from the heart. However, in order to accurately segment the substructure of the heart, specialized knowledge of cardiac anatomy is required, and depending on the expert's processing, there may be a problem in which the left ventricular EF is calculated differently. In this study, using the DeepLabV3 architecture, GBP images were trained on 93 training data with a ResNet-50 backbone. Afterwards, the trained model was applied to 23 separate test sets of GBP to evaluate the reproducibility of the region of interest and left ventricular EF. Pixel accuracy, dice coefficient, and IoU for the region of interest were 99.32±0.20, 94.65±1.45, 89.89±2.62(%) at the diastolic phase, and 99.26±0.34, 90.16±4.19, and 82.33±6.69(%) at the systolic phase, respectively. Left ventricular EF was calculated to be an average of 60.37±7.32% in the ROI set by humans and 58.68±7.22% in the ROI set by the deep learning segmentation model. (p<0.05) The automated segmentation method using deep learning presented in this study similarly predicts the average human-set ROI and left ventricular EF when a random GBP image is an input. If the automatic segmentation method is developed and applied to the functional examination method that needs to set ROI in the field of cardiac scintigram in nuclear medicine in the future, it is expected to greatly contribute to improving the efficiency and accuracy of processing and analysis by nuclear medicine specialists.

• Journal of Platform Technology
• /
• v.9 no.4
• /
• pp.71-78
• /
• 2021

Self-driving vehicles are increasing as new alternatives to solving problems such as human safety, environment and aging. And such technology development has a great ripple effect on other industries. However, various problems are occurring. The number of casualties caused by self-driving is increasing. Although the collision of fixed obstacles is somewhat decreasing, on the contrary, the technology by active obstacles is still insignificant. Therefore, in this study, in order to solve the core problem of self-driving vehicles, we propose a method of extracting active obstacles on the road. First, a center scene is extracted from a continuous image. In addition, it was proposed to extract activity obstacles using activity size and activity repeatability information from objects included in the center scene. The center scene is calculated using region segmentation and merging. Based on these results, the size of the frequency for each pixel in the region was calculated and the size of the activity of the obstacle was calculated using information that frequently appears in activity. Compared to the results extracted directly by humans, the extraction accuracy was somewhat lower, but satisfactory results were obtained. Therefore, it is believed that the proposed method will contribute to solving the problems of self-driving and reducing human accidents.

• Journal of Korea Society of Industrial Information Systems
• /
• v.27 no.4
• /
• pp.1-10
• /
• 2022

Attempts to interpret human perspectives using computer vision have been developed in various fields. In this paper, we propose a method for evaluating the walking environment through semantic segmentation results of images from road images. First, the Kakao Map API was used to collect road images, and four-way images were collected from about 50,000 points in JeonJu. 20% of the collected images build datasets through crowdsourcing-based paired comparisons, and train various regression models using paired comparison data. In order to derive the walkability score of the image data, the ranking score is calculated using the Trueskill algorithm, which is a ranking algorithm, and the walkability and analysis using various regression models are performed using the constructed data. Through this study, it is shown that the walkability of Jeonju can be evaluated and scores can be derived through the correlation between pixel distribution classification information rather than human vision.

• Journal of Advanced Navigation Technology
• /
• v.26 no.2
• /
• pp.113-118
• /
• 2022

Motivated by people highlighting important phrases while reading or taking notes we propose a neural network training method by highlighting defective pixel areas to classify effectively defect types of images with complex background textures. To verify our proposed method we apply it to the problem of classifying the defect types of tire band fabric images that are too difficult to classify. In addition we propose a backlight highlighting technique which is tailored to the tire band fabric images. Backlight highlighting images can be generated by using both the GradCAM and simple image processing. In our experiment we demonstrated that the proposed highlighting method outperforms the traditional method in the view points of both classification accuracy and training speed. It achieved up to 13.4% accuracy improvement compared to the conventional method. We also showed that the backlight highlighting technique tailored for highlighting tire band fabric images is superior to a contour highlighting technique in terms of accuracy.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.3
• /
• pp.211-216
• /
• 2022

Steganography technology protects the existence of hidden information by embedding a secret message in a specific location on the cover medium. Security and resistance are strengthened by applying various hybrid methods based on encryption and steganography. In particular, techniques to increase chaos and randomness are needed to improve security. In fact, the case where the shuffling method is applied based on the discrete cosine transform(DCT) and the least significant bit(LSB) is an area that needs to be studied. I propose a new approach to hide the bit information of Hangul messages by integrating the selective shuffling method that can add the complexity of message hiding and applying the spatial domain technique to steganography. Inverse shuffling is applied when extracting messages. In this paper, the Hangul message to be inserted is decomposed into the choseong, jungseong and jongseong. It improves security and chaos by applying a selective shuffling process based on the corresponding information. The correlation coefficient and PSNR were used to confirm the performance of the proposed method. It was confirmed that the PSNR value of the proposed method was appropriate when compared with the reference value.