• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1331-1344
• /
• 2017

In this paper, data hiding algorithm using Discrete Wavelet Transform (DWT) and Arnold Transform is proposed. The secret data is scrambled using Arnold Transform to make it secure. Wavelet subbands of a cover image are obtained using DWT. The scrambled secret data is embedded into significant wavelet coefficients of subbands of a cover image. The proposed algorithm is robust to a variety of attacks like JPEG and JPEG2000 compression, image cropping and median filtering. Experimental results show that the PSNR of the composite image is 1.05 dB higher than the PSNR of existing algorithms and capacity is 25% higher than the capacity of existing algorithms.

• 한국멀티미디어학회논문지
• /
• 제23권7호
• /
• pp.812-818
• /
• 2020

Cervical cancer is the second most common female cancer in the world. In Korea, cervical cancer accounts for 13 percent of female cancers and 4,200 cases occur annually[1]. The purpose of this study is to use a deep learning model to identify the possibility of lesions in the cervix and to evaluate the efficient image preprocessing in order to diagnose diverse types of cervix in form. The study used 4,107 normal photographs of uterine cervix and 6,285 abnormal photographs of uterine cervix. Two types of image preprocessing were resized to square. The methods are cropping based on height and filling the space up and down with black images. In addition, all images were resampled to 256×256. The average accuracy of cropped cases is 94.15%. The average accuracy of the filled cases is 93.41%. According to the study, the model performance of cropped data was slightly better. But there were several images that were not accurately classified. Therefore, the additional experiment with pre-treatment process based on cropping is needed to cover images of the cervix in more detail.

• 대한원격탐사학회지
• /
• 제20권3호
• /
• pp.175-188
• /
• 2004

The objectives of this study were to estimate leaf area index (LAI) as a function of image-derived vegetation indices, and to compare measured and estimated LAI to the results of crop model simulation. Soil moisture, crop phenology, and LAI data were obtained several times during the 2001 growing season at monitoring sites established in two central Missouri experimental fields, one planted to com (Zea mays L.) and the other planted to soybean (Glycine max L.). Hyper- and multi-spectral images at varying spatial. and spectral resolutions were acquired from both airborne and satellite platforms, and data were extracted to calculate standard vegetative indices (normalized difference vegetative index, NDVI; ratio vegetative index, RVI; and soil-adjusted vegetative index, SAVI). When comparing these three indices, regressions for measured LAI were of similar quality $(r^2$ =0.59 to 0.61 for com; $r^2$ =0.66 to 0.68 for soybean) in this single-year dataset. CERES(Crop Environment Resource Synthesis)-Maize and CROPGRO-Soybean models were calibrated to measured soil moisture and yield data and used to simulate LAI over the growing season. The CERES-Maize model over-predicted LAI at all corn monitoring sites. Simulated LAI from CROPGRO-Soybean was similar to observed and image-estimated LA! for most soybean monitoring sites. These results suggest crop growth model predictions might be improved by incorporating image-estimated LAI. Greater improvements might be expected with com than with soybean.

• 한국멀티미디어학회논문지
• /
• 제15권7호
• /
• pp.891-902
• /
• 2012

Reversible data hiding scheme is a form of steganography in which the secret embedding data can be retrieved from a stego image for the purpose of identification, copyright protection and making a covert channel. The reversible data hiding should satisfy that not only are the distortions due to artifacts against the cover image invisible but also it has large embedding capacity as far as possible. In this paper, we propose a robust reversible data hiding scheme by exploiting the differences between a center pixel and its neighboring pixels in each sub-block of the image to embed secret data into extra space. Moreover, our scheme enhances the embedding capacity and can recover the embedded data from the stego image without causing any perceptible distortions to the cover image. Simulation results show that our proposed scheme has lower visible distortions in the stego image and provides robustness to geometrical image manipulations, such as rotation and cropping operations.

• 한국컴퓨터정보학회논문지
• /
• 제24권1호
• /
• pp.9-23
• /
• 2019

In this paper, we propose a layer structure of a pest image classifier model using CNN (Convolutional Neural Network) and background removal image processing algorithm for improving classification accuracy in order to build a smart monitoring system for pine wilt pest control. In this study, we have constructed and trained a CNN classifier model by collecting image data of pine wilt pest mediators, and experimented to verify the classification accuracy of the model and the effect of the proposed classification algorithm. Experimental results showed that the proposed method successfully detected and preprocessed the region of the object accurately for all the test images, resulting in showing classification accuracy of about 98.91%. This study shows that the layer structure of the proposed CNN classifier model classified the targeted pest image effectively in various environments. In the field test using the Smart Trap for capturing the pine wilt pest mediators, the proposed classification algorithm is effective in the real environment, showing a classification accuracy of 88.25%, which is improved by about 8.12% according to whether the image cropping preprocessing is performed. Ultimately, we will proceed with procedures to apply the techniques and verify the functionality to field tests on various sites.

• 방송공학회논문지
• /
• 제7권2호
• /
• pp.192-201
• /
• 2002

This paper develops a novel robust information hiding technique that uses channel codes derived from the error-correcting coder. The message encoded by the cover encoder is hidden in DCT transform domain of the cover image. The method exploits the sensitivity of human eyes to adaptively embed a visually recognizable message in an image without affecting the perceptual quality of the underlying cover image. Experimental results show that the proposed data hiding technique is robust to cropping operations, lossy JPEG compression, noise interference and secure against known stego attacks. The performance of the proposed scheme with turbo coder is superior to that without turbo coder.

• 전기학회논문지
• /
• 제67권5호
• /
• pp.669-674
• /
• 2018

Data augmentation is an efficient way to reduce overfitting on models and to improve a performance supplementing extra data for training. It is more important in deep learning based industrial machine vision. Because deep learning requires huge scale of learning data to learn a model, but acquisition of data can be limited in most of industrial applications. A very generic method for augmenting image data is to perform geometric transformations, such as cropping, rotating, translating and adjusting brightness of the image. The effectiveness of data augmentation in image classification has been reported, but it is rare in defect inspections. We explore and compare various basic augmenting operations for the metal surface defects. The experiments were executed for various types of defects and different CNN networks and analysed for performance improvements by the data augmentations.

• 전자공학회논문지CI
• /
• 제41권6호
• /
• pp.101-109
• /
• 2004

본 논문에서는 인간의 시각적 특성에 기반한 적응적 워터마킹 알고리즘을 제안한다. 워터마크를 삽입할 대상에 대한 시각적 특성 분석을 위해 영상의 휘도, 대비, 텍스쳐 등의 특징을 사용하였다. 제안하는 방법은 인간의 시각적 특성에 따라 워터마크의 삽입 블록과 삽입될 워터마크의 위치와 강도를 적응적으로 결정한다. 실험은 절단, 영상강화, 저주파 필터링, JPEG 압축의 영상처리 공격에 대한 워터마크 검출율을 비교하였다.

• 대한임베디드공학회논문지
• /
• 제16권6호
• /
• pp.277-283
• /
• 2021

This paper addresses the problem of vision-based mask filter inspection for mask production systems. Machine learning-based approaches can be considered to solve the problem, but they may not be applicable to mask filter inspection if normal and anomaly mask filter data are not sufficient. In such cases, handcrafted image processing methods have to be considered to solve the problem. In this paper, we propose a hierarchical correlation-based approach that combines handcrafted image processing methods to detect anomaly mask filters. The proposed approach combines image rotation, cropping and resizing, edge detection of mask filter parts, average blurring, and correlation-based decision. The proposed approach was tested and analyzed with real mask filters. The results showed that the proposed approach was able to successfully detect anomalies in mask filters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권9호
• /
• pp.4405-4418
• /
• 2016

Automatic image completion techniques have difficulty processing images in which the target region has multiple planes or is non-facade. Here, we propose a new image completion method that uses belief propagation based on planar priorities. We first calculate planar information, which includes planar projection parameters, plane segments, and repetitive regularity extractions within the plane. Next, we convert this planar information into planar guide knowledge using the prior probabilities of patch transforms and offsets. Using the energy of the discrete Markov Random Field (MRF), we then define an objective function for image completion that uses the planar guide knowledge. Finally, in order to effectively optimize the MRF, we propose a new optimization scheme, termed Planar Priority-belief propagation that includes message-scheduling-based planar priority and dynamic label cropping. The results of experiment show that our approach exhibits advanced performance compared with existing approaches.