• Journal of information and communication convergence engineering
• /
• v.19 no.2
• /
• pp.79-83
• /
• 2021

This paper proposes a novel image classification method based on few-shot learning, which is mainly used to solve model overfitting and non-convergence in image classification tasks of small datasets and improve the accuracy of classification. This method uses model structure optimization to extend the basic convolutional neural network (CNN) model and extracts more image features by adding convolutional layers, thereby improving the classification accuracy. We incorporated certain measures to improve the performance of the model. First, we used general methods such as setting a lower learning rate and shuffling to promote the rapid convergence of the model. Second, we used the data expansion technology to preprocess small datasets to increase the number of training data sets and suppress over-fitting. We applied the model to 10 monkey species and achieved outstanding performances. Experiments indicated that our proposed method achieved an accuracy of 87.92%, which is 26.1% higher than that of the traditional CNN method and 1.1% higher than that of the deep convolutional neural network ResNet50.

• Journal of Korean Society for Quality Management
• /
• v.51 no.1
• /
• pp.119-129
• /
• 2023

Purpose: This study proposes a classification model for implementing condition-based maintenance (CBM) by monitoring the real-time status of a machine using acceleration sensor data collected from a vehicle. Methods: The classification model's performance was improved by applying Fourier transform to convert the acceleration sensor data from the time domain to the frequency domain. Additionally, the Generative Adversarial Network (GAN) algorithm was used to augment images and further enhance the classification model's performance. Results: Experimental results demonstrate that the GAN algorithm can effectively serve as an image augmentation technique to enhance the performance of the classification model. Consequently, the proposed approach yielded a significant improvement in the classification model's accuracy. Conclusion: While this study focused on the effectiveness of the GAN algorithm as an image augmentation method, further research is necessary to compare its performance with other image augmentation techniques. Additionally, it is essential to consider the potential for performance degradation due to class imbalance and conduct follow-up studies to address this issue.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• 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.

• Korean Journal of Remote Sensing
• /
• v.30 no.1
• /
• pp.25-36
• /
• 2014

Multispectral image classification of individual tree species is often difficult because of the spectral similarity among species. In this study, we attempted to analyze the suitability of hyperspectral image to classify coniferous tree species. Several image sets and classification methods were applied and the classification results were compared with the ones from multispectral image. Two airborne hyperspectral images (AISA, CASI) were obtained over the study area in the Gwangneung National Forest. For the comparison, ETM+ multispectral image was simulated using hyperspectral images as to have lower spectral resolution. We also used the transformed hyperspectral data to reduce the data volume for the classification. Three supervised classification schemes (SAM, SVM, MLC) were applied to thirteen image sets. In overall, hyperspectral image provides higher accuracies than multispectral image to discriminate coniferous species. AISA-dual image, which include additional SWIR spectral bands, shows the best result as compared with other hyperspectral images that include only visible and NIR bands. Furthermore, MNF transformed hyperspectral image provided higher classification accuracies than the full-band and other band reduced data. Among three classifiers, MLC showed higher classification accuracy than SAM and SVM classifiers.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.3
• /
• pp.1-9
• /
• 2021

This research examines deep learning based image recognition models for beef sirloin classification. The sirloin of beef can be classified as the upper sirloin, the lower sirloin, and the ribeye, whereas during the distribution process they are often simply unified into the sirloin region. In this work, for detailed classification of beef sirloin regions we develop a model that can learn image information in a reasonable computation time using the MobileNet algorithm. In addition, to increase the accuracy of the model we introduce data augmentation methods as well, which amplifies the image data collected during the distribution process. This data augmentation enables to consider a larger size of training data set by which the accuracy of the model can be significantly improved. The data generated during the data proliferation process was tested using the MobileNet algorithm, where the test data set was obtained from the distribution processes in the real-world practice. Through the computational experiences we confirm that the accuracy of the suggested model is up to 83%. We expect that the classification model of this study can contribute to providing a more accurate and detailed information exchange between suppliers and consumers during the distribution process of beef sirloin.

• Korean Journal of Remote Sensing
• /
• v.27 no.6
• /
• pp.637-651
• /
• 2011

This study presents a rule-based urban image classification method for time series analysis of changes in the vicinity of Asan-si and Cheonan-si in Chungcheongnam-do, using Landsat satellite images (1991-2006). The area has been highly developed through the relocation of industrial facilities, land development, construction of a high-speed railroad, and an extension of the subway. To determine the yearly changing pattern of the urban area, eleven classes were made depending on the trend of development. An algorithm was generalized for the rules to be applied as an unsupervised classification, without the need of training area. The analysis results show that the urban zone of the research area has increased by about 1.53 times, and each correlation graph confirmed the distribution of the Built Up Index (BUI) values for each class. To evaluate the rule-based classification, coverage and accuracy were assessed. When Optimal allowable factor=0.36, the coverage of the rule was 98.4%, and for the test using ground data from 1991 to 2006, overall accuracy was 99.49%. It was confirmed that the method suggested to determine the maximum allowable factor correlates to the accuracy test results using ground data. Among the multiple images, available data was used as best as possible and classification accuracy could be improved since optimal classification to suit objectives was possible. The rule-based urban image classification method is expected to be applied to time series image analyses such as thematic mapping for urban development, urban development, and monitoring of environmental changes.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.5 no.2
• /
• pp.24-33
• /
• 1987

This study aims at obtaining more effective image processing techniques and more accurately classified image in the sphere which uses remotely sensed data. For this practice, the result of land use classification compounding summer scene with winter scene and the classified result of summer scene were compared, analyzed. From the upper analysed results, we found that Bitemporal Classification technique and $tan^{-1}$transformation were effective. Particularly, dividing crop class into two classes of farmland and field was more possible by appling Bitemporal Classification technique.

• Korean Journal of Remote Sensing
• /
• v.22 no.2
• /
• pp.101-110
• /
• 2006

The accuracy of rural land cover using airborne multispectral images and LEAR (Light Detection And Ranging) data was analyzed. Multispectral image consists of three bands in green, red and near infrared. Intensity image was derived from the first returns of LIDAR, and vegetation height image was calculated by difference between elevation of the first returns and DEM (Digital Elevation Model) derived from the last returns of LIDAR. Using maximum likelihood classification method, three bands of multispectral images, LIDAR vegetation height image, and intensity image were employed for land cover classification. Overall accuracy of classification using all the five images was improved to 85.6% about 10% higher than that using only the three bands of multispectral images. The classification accuracy of rural land cover map using multispectral images and LIDAR images, was improved with clear difference between heights of different crops and between heights of crop and tree by LIDAR data and use of LIDAR intensity for land cover classification.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.2
• /
• pp.33-38
• /
• 2020

Anomaly detection of Machine Learning such as PCA anomaly detection and CNN image classification has been focused on cross-sectional data. In this paper, two approaches has been suggested to apply ML techniques for identifying the failure time of big time series data. PCA anomaly detection to identify time rows as normal or abnormal was suggested by converting subjects identification problem to time domain. CNN image classification was suggested to identify the failure time by re-structuring of time series data, which computed the correlation matrix of one minute data and converted to tiff image format. Also, LASSO, one of feature selection methods, was applied to select the most affecting variables which could identify the failure status. For the empirical study, time series data was collected in seconds from a power generator of 214 components for 25 minutes including 20 minutes before the failure time. The failure time was predicted and detected 9 minutes 17 seconds before the failure time by PCA anomaly detection, but was not detected by the combination of LASSO and PCA because the target variable was binary variable which was assigned on the base of the failure time. CNN image classification with the train data of 10 normal status image and 5 failure status images detected just one minute before.

• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.1
• /
• pp.121-128
• /
• 2016

Many studies are focused on image data and classifier for comparison or improvement of classification accuracy. Therefore studies are needed aspect of the training samples on supervised classification which depend on reference data or skill of analyst. This study tries to assess usability of field spectra as training samples on supervised classification. Classification accuracies of hyperspectral and multispectral images were assessed using training samples from image itself and field spectra, respectively. The results shown about 90% accuracy with training sample collected from image. Using field spectra as training sample, accuracy was decreased 10%p for hyperspectral image, and 20%p for multispectral image. Especially, some classes shown very low accuracies due to similar spectral characteristics on multispectral image. Therefore, field spectra might be used as training samples on classification of hyperspectral image, although it has limitation for multispectral image.