Bachelor Degree [September 2005 - August 2009]

Faculty of Computer Science, Universitas Indonesia Depok, Indonesia.

Thesis Title : Development of a Dimension-Based Learning Algorithm and Its Comparison with Vector-Based Learning using Fuzzy-Neuro Learning Vector Quantization to Recognize Frontal Face Images

Qualifications received : Sarjana Ilmu Komputer (Bachelor in Computer Science)

• R. Maulana, M. F. Rachmadi, L. Rahadianti. Robustness of Probabilistic U-Net for Automated Segmentation of White Matter Hyperintensities in Different Datasets of Brain MRI. Proceedings of the International Conference on Advanced Computer Science and Information Systems, 2021.

• S. Nurhayati, L. Rahadianti, D. Chahyati, P. A. Yusuf, E.D. Tenda, R.E. Yunus. Chest X-Ray Patch Classification for Tuberculosis Detection. Proceedings of the International Conference on Advanced Computer Science and Information Systems, 2021.

• F.H. Muttaqien, L. Rahadianti, A. L. Latifah. Downscaling for Climate Data in Indonesia Using Image-to-Image Translation Approach. Proceedings of the International Conference on Advanced Computer Science and Information Systems, 2021.

• F. H. Caryn, L. Rahadianti. "Driver Drowsiness Detection Based on Drivers’ Physical Behaviours: A Systematic Literature Review". Computer Engineering and Applications Journal, Vol 10, No. 3 (2021).

• L. Rahadianti, A.Y. Azizah, and H. Deborah. "Evaluation of the Quality Indicators in Dehazed Images: Color, Contrast, Naturalness, and Visual Pleasingness". Heliyon (2021): e08038.

• E. Yulianti, L. Rahadianti. "Determining subject headings of documents using information retrieval models". Indonesian Journal of Electrical Engineering and Computer Science, Vol 23, No 2 (2021).

• C.A. Hartanto, L. Rahadianti. "Single Image Dehazing Using Deep Learning." JOIV: International Journal on Informatics Visualization 5.1 (2021): 76-82.

• Rahadianti, L. "3D Information from Scattering Media Images". Jurnal Ilmu Komputer dan Informasi 14, no. 1 (2021): 73-82.

• A.Y. Azizah, L. Rahadianti, H. Deborah. An Introductory Study on Image Quality of Dehazed Images. Proceedings of the International Conference on Advanced Computer Science and Information Systems, 2020.

• M. Ramadiansyah, L. Rahadianti. Proxy-Based and Pair-Based Losses for Face Image Retrieval. Proceedings of the International Conference on Advanced Computer Science and Information Systems, 2020.

• L. Rahadianti, F. Sakaue, J. Sato. Depth Estimation from Single Hazy Images with 2-Phase Training. Proceedings of the International Conference on Advanced Computer Science and Information Systems, 2020.

• L. Rahadianti, F. Sakaue, J. Sato, Depth Estimation for Hazy Images using Deep Learning, In Proceedings of the 4th Asian Conference on Pattern Recognition, pp.238-243, 2017.

• L. Rahadianti, F. Sakaue, J. Sato. Time-to-contact in Scattering Media Environments based on Statistical Priors. ITE Transactions on Media Technology and Applications, Vol. 5 no.4, pp. 147-161, 2017.

• L. Rahadianti, W. Jeong, F. Sakaue, J. Sato. Time-to-Contact in Scattering Media. IEICE Transcations on Information and Systems, Vol 100-D no. 3, pp. 564--573, 2017.

• L. Rahadianti, F. Sakaue, J. Sato. Time-to-Contact from Underwater Images. In Proceedings of the 11th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP). Rome, Italy, pp. 673--680, 2016.

• W. Jeong, L. Rahadianti, F. Sakaue, J. Sato. Time-to-Contact in Scattering Media. In Proceedings of the 10th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISAPP). Berlin, Germany, pp. 658--663, 2015.

• A.S. Imran, L. Rahadianti, F.A. Cheikh, S.Y. Yayilgan. Objective Keyword Selection for Lecture Video Annotation. In Proceedings of the 5th European Workshop in Visual Information Processing (EUVIP). Paris, France, 2014.

• A.S. Imran, L. Rahadianti, F.A. Cheikh, S.Y. Yayilgan. Semantic Keyword Selection for Automatic Video Annotation. In Proceedings of the International Conference on Signal-Image Technology and Internet-Based Systems (SITIS). Kyoto, Japan, pp. 241--246, 2013.

• A.S. Imran, L. Rahadianti, F.A. Cheikh, S.Y. Yayilgan. Semantic Tags for Lecture Videos. In Proceedings of the 6th International Conference on Semantic Computing (ICSC). Palermo, Italy, pp.117--120, 2012.

• L. Rahadianti, H.R. Sanabila, R. Manurung, and A. Murni. Batik Image Classification using Log-Gabor and Generalized Hough Transform Features. In Proceedings of the 2nd International Conference on Soft Computing, Intelligent System, and Information Technology (ICSIIT). Bali, Indonesia, pp.320-325, 2010.

• L. Rahadianti, R. Manurung, and A. Murni. Clustering Batik Images based on Log-Gabor and Colour Histogram Features . In Proceedings of the International Conference on Advanced Computer Science and Information Systems (ICACSIS). Depok, December 7-8, 2009.

• Erasmus Mundus Scholarship [September 2010 - August 2012]

Erasmus Mundus Scholarship funded by the European Union for post graduate degrees in Erasmus Mundus Joint Master Degree Programs in Europe. (Subsequently renamed as Erasmus+ programs). Full scholarship awarded for the 2 year master study period in Erasmus Mundus Color in Informatics and Media Technology (CIMET).

Pre-processing Images in the Wild for OCR

Optical Character Recognition (OCR) is useful to extract text information from images, and many tools are available for this. However, in order to use OCR tools, images must be in an optimal condition to be processed. Images in the wild must be pre-processed in preparation of OCR. Thus, we study the case of various types of images in the wild to propose a framework of pre-processing.

• • Over-the-counter vitamin packaging recognition

  • Lecture video processing for OCR

This work is openly recruiting students, and involves collaboration with the Information Retrieval-Natural Language Processing (IR-NLP) Lab.

Deep Learning for Time Series Images

In recent years, deep learning has become a proven tool to solve multidisciplinary problems, including image to image translation. Aside from single images, it is also possible to consider images in a time sequence, from which serious information can be obtained based on the order that the images are captured. This research will attempt to tackle various time-series image data to perform the real-world tasks related to them, such as those above, using deep learning networks.

• • Statistical Downscaling for Temporal Weather Data

  • Action Recognition from Video

  • Metric Learning for Image Representation

Includes the supervision of 1 (current) student and 2 graduates in the Master of Computer Science program, Faculty of Computer Science, Universitas Indonesia. This work was supported by internal research funding from UI.

Image Processing in Scattering Media

Computer vision applications attempt to understand a scene from captured images, however most works assume that the image is captured in clear media. However, in many real-world conditions, such as foggy or underwater environments, the surrounding media may contain microparticles that affect light propagation. This type of media is called scattering media, and the nature of it makes 3D scene understanding using conventional methods difficult. This research will attempt to model the photometric properties of images in scattering media as well as explore deep learning approaches to enable automated 3D scene understanding of single still images in scattering media.

• • Image Restoration of Scattering Media

  • 3D Distance Estimation from Images in Scattering Media

Includes the supervision of 2 (current) students and 1 graduate in the Master of Computer Science program; and 2 (current) students and 5 graduates of the Bachelor of Computer Science program, Faculty of Computer Science, Universitas Indonesia. This work was supported by internal research funding from UI.

Image Quality Analysis

The quality of restored images are often only evaluated by how well it can recover the scene, measured using a metric such as PSNR, etc. Meanwhile, as a general image quality assessment problem, it is necessary to measure how visually pleasing the image is. Considering image quality, many factors of human perception come into play. Many of these factors can not be measured by the metrics commonly used. The thorough assessment of image quality is a holistic approach to the image appearance as a whole, not only with one or a few metrics, with specific image quality indicators for each method. This can be especially useful since ground truth is not always available.

• Image Quality Assessment of Dehazed Images

• Image Quality Metrics for Subjective and Objective Assessment

This work is openly recruiting students, and includes the supervision of 1 (current) student of the Bachelor of Computer Science program, Faculty of Computer Science, Universitas Indonesia. This project involves collaboration with NTNU - Norwegian University of Science and Technology.

Multi and Hyper-Spectral Image Processing

The digital color images that are commonly known in daily life are captured with 3 bands, R, G, and B. However, recall that images are sensed on the electromagnetic spectrum, allowing image capture along it at various sampled wavelengths. In multispectral imaging, color can be represented by a fuller spectrum to give more information about the color. This can help the differentiation of colors that seem equal to the naked eye (which senses images on 3 bands as well), among other applications.

• The analysis of the effect of haze on multi-spectral images

• Ink clustering using multi-spectral images

Including the supervision of 2 graduates of the Bachelor of Computer Science program, Faculty of Computer Science, Universitas Indonesia. This project involves collaboration with NTNU - Norwegian University of Science and Technology.

Image Processing for Color Vision Deficiency

Color is a sensation, which can be perceived by human observers thanks to the 3 photoreceptors (cones) in the retina. The spectral sensitivity of the 3 cones are sensitive to the R, G, and B colors on the electromagnetic spectrum. However, in some humans, the spectral sensitivity of the photoreceptors stray from the normal sensitivities. These abnormal sensitivities result in Color Vision Deficiency, in which the observer is not able to perceive color correctly. These anomalies can occur in any of the R, G, and B spectral sensitivities, and they can either be weakened or absent completely. Since this sensing occurs in the eye and in the signal the eye sends to the brain, it is impossible to change the perception of the color itself. However, it is possible to process color images in such a way that it is easier to understand. Daltonization, recoloring, and other color correction techniques may be used to adjust colors tp be less confusing to a color blind person.

Including the supervision of 1 (current) student and 1 graduate of the Bachelor of Computer Science program, Faculty of Computer Science, Universitas Indonesia.

AI for TB Detection

Tuberculosis (TB) still becomes a burden in global health with one in four people in the world being infected with TB bacteria. Artificial intelligence (AI) approaches can support physician diagnosis using automated digital medical imaging processing. This work will attempt to process CXR imaging using artificial intelligence processes to help physicians to detect tuberculosis in patients.

Including the supervision of 1 graduate of the Master of Computer Science program, Faculty of Computer Science, Universitas Indonesia. This project involves collaboration with the Indonesia Medical Education and Research Institute (IMERI). Faculty of Medicine, Universitas Indonesia. This work is supported by internal research funding from UI.

Digital Stethoscope

Tuberculosis (TB) still becomes a burden in global health with one in four people in the world being infected with TB bacteria. Due to the COVID-19 pandemic, less patients make their way physically to a health facility in which the gold standard BTA culture test can be done to diagnose TB. Thus, we investigate alternate methods to establish a computer-aided-diagnosis of TB, in this case a commercial digital stethoscope. The recording of the auscultation of lungs using the digital stethoscope will be processed by a automatic classifier using AI methods.

This work is openly recruiting students, and involves collaboration with the Indonesia Medical Education and Research Institute (IMERI). Faculty of Medicine, Universitas Indonesia.