Aplikasi Filter Digital untuk Citra Mengandung Derau Salt and Pepper Menggunakan Metode Directional Weighted Minimum Deviation

Teknologi telekomunikasi berkembang begitu pesat, dari yang semula berkomunikasi menggunakan surat, berkembang komunikasi suara menggunakan telepon hingga kini telah sampai pada tahap komunikasi gambar dan video. Dalam proses pentransmisian data baik suara maupun gambar tidak terlepas dari adanya derau. Salah satu solusi dalam menjawab permasalahan tersebut adalah dengan mengembangkan teknologi Filter Digital. Dalam penelitian ini direalisasikan sebuah Filter digital dengan objek gambar menggunakan metode DWMD (Directionan Weighted Minumum Deviation) Filter dengan mendeteksi jenis derau salt and pepper. Metode DWMD Filter adalah metode pengolahan data digital berbasis arah dan standar deviasi yang memperbaiki Median Filter. Dengan membandingkan parameter PSNR maka diketahui bahwa DWMD dapat menghasilkan gambar lebih baik dari Median Filter. Metode Filter DWMD ini ditambah dengan penentuan Threshold otomatis untuk membantu proses filter menjadi lebih cepat dimana diambil selisih PSNR terbesar. Hasil eksperimen menunjukkan bahwa pada level derau 5% - 65% metode DWMD menghasilkan PSNR bernilai 21-36 dB dibandingkan dengan Median Filter sebesar 13-29 dB. Penelitian ini memiliki output berupa aplikasi desktop yang dilengkapi dengan fitur-fitur yang dapat menambahkan derau salt and pepper pada berbagai densitas dan mengakses citra yang dapat diubah ke dalam format grayscale.

Fruit-classification model resilience under adversarial attack

An accurate and robust fruit image classifier can have a variety of real-life and industrial applications including automated pricing, intelligent sorting, and information extraction. This paper demonstrates how adversarial training can enhance the robustness of fruit image classifiers. In the past, research in deep-learning-based fruit image classification has focused solely on attaining the highest possible accuracy of the model used in the classification process. However, even the highest accuracy models are still susceptible to adversarial attacks which pose serious problems for such systems in practice. As a robust fruit classifier can only be developed with the aid of a fruit image dataset consisting of fruit images photographed in realistic settings (rather than images taken in controlled laboratory settings), a new dataset of over three thousand fruit images belonging to seven fruit classes is presented. Each image is carefully selected so that its classification poses a significant challenge for the proposed classifiers. Three Convolutional Neural Network (CNN)-based classifiers are suggested: 1) IndusNet, 2) fine-tuned VGG16, and 3) fine-tuned MobileNet. Fine-tuned VGG16 produced the best test set accuracy of 94.82% compared to the 92.32% and the 94.28% produced by the other two models, respectively. Fine-tuned MobileNet has proved to be the most efficient model with a test time of 9 ms/step compared to the test times of 28 ms/step and 29 ms/step for the other two models. The empirical evidence presented demonstrates that adversarial training enables fruit image classifiers to resist attacks crafted through the Fast Gradient Sign Method (FGSM), while simultaneously improving classifiers’ robustness against other noise forms including ‘Gaussian’, ‘Salt and pepper’ and ‘Speckle’. For example, when the amplitude of the perturbations generated through the Fast Gradient Sign Method (FGSM) was kept at 0.1, adversarial training improved the fine-tuned VGG16’s performance on adversarial images by around 18% (i.e., from 76.6% to 94.82%), while simultaneously improving the classifier’s performance on fruit images corrupted with ‘salt and pepper’ noise by around 8% (i.e., from 69.82% to 77.85%). Other reported results also follow this pattern and demonstrate the effectiveness of adversarial training as a means of enhancing the robustness of fruit image classifiers.

Dynamic monitoring of urban built-up object expansion trajectories in Karachi, Pakistan with time series images and the LandTrendr algorithm

In the complex process of urbanization, retrieving its dynamic expansion trajectories with an efficient method is challenging, especially for urban regions that are not clearly distinguished from the surroundings in arid regions. In this study, we propose a framework for extracting spatiotemporal change information on urban disturbances. First, the urban built-up object areas in 2000 and 2020 were obtained using object-oriented segmentation method. Second, we applied LandTrendr (LT) algorithm and multiple bands/indices to extract annual spatiotemporal information. This process was implemented effectively with the support of the cloud computing platform of Earth Observation big data. The overall accuracy of time information extraction, the kappa coefficient, and average detection error were 83.76%, 0.79, and 0.57 a, respectively. These results show that Karachi expanded continuously during 2000–2020, with an average annual growth rate of 4.7%. However, this expansion was not spatiotemporally balanced. The coastal area developed quickly within a shorter duration, whereas the main newly added urban regions locate in the northern and eastern inland areas. This study demonstrated an effective framework for extract the dynamic spatiotemporal change information of urban built-up objects and substantially eliminate the salt-and-pepper effect based on pixel detection. Methods used in our study are of general promotion significance in the monitoring of other disturbances caused by natural or human activities.