Search

Exploiting sensitive human data in human visual monitoring system violates individual’s privacy. Hence, this study utilized a Light Detection and Ranging (LiDAR)-generated three-dimensional (3D) point cloud instead of an RGB camera as it captures the human object without detailed imagery. Given their dispersed nature, processing 3D LiDAR point cloud is economically inefficient as it requires some preliminary actions. Alternatively, this study applied a single-stage detection process using only one data type, namely 3D LiDAR point cloud with vertical axis direct projection. This approach utilized the 3D LiDAR cylindrical coordinates features to project the data onto two-dimensional (2D) image space with various computational capabilities and back-projection algorithm to restore the identified object on the 2D plane onto 3D LiDAR point cloud coordinates. This model also implemented the YOLOv5 series due to their varied sizes. The evaluation of this approach utilized accuracy metric which was the average of mean average precision (mAP) value based on different intersections over union (IoU) thresholds. The proposed methodology effectively employed a vertical projection technique to identify human objects. Notably, this approach distinguishes itself from previous methods such as PIXOR, BirdNet, BirdNet+, BEVDetNet, and Frustum-PointPillars, offering a novel perspective in the field. In addition, the best and worst performing models had the accuracy values of 44.35% and 79.83%, with inference speeds of 3.7 ms and up to 25 ms, respectively. Further, the inference speeds of all models were less than 33.33 ms. Thus, the monitored objects were identified before the LiDAR system enters the next azimuth rotation.

Traditional dances, like those of Indonesia, have complex and unique patterns requiring accurate cultural preservation and documentation classification. However, traditional dance classification methods often rely on manual analysis and subjective judgment, which leads to inconsistencies and limitations. This research explores a modified deep pattern classifier of traditional dance movements in videos, including Gambyong, Remo, and Topeng, using a Convolutional Neural Network (CNN). Evaluation model's performance using a testing spatio-temporal dataset in Indonesian traditional dance videos is performed. The videos are processed through frame-level segmentation, enabling the CNN to capture nuances in posture, footwork, and facial expressions exhibited by dancers. Then, the obtained confusion matrix enables the calculation of performance metrics such as accuracy, precision, sensitivity, and F1-score. The results showcase a high accuracy of 97.5%, indicating the reliable classification of the dataset. Furthermore, future research directions are suggested, including investigating advanced CNN architectures, incorporating temporal information through recurrent neural networks, exploring transfer learning techniques, and integrating user feedback for iterative refinement of the model. The proposed method has the potential to advance dance analysis and find applications in dance education, choreography, and cultural preservation.

ABSTRAK Pupilometri merupakan metode pengukuran respons pupil terhadap stimulus. Kemampuan pupil mata dalam merespons cahaya diamati melalui pupillary light response (PLR). Penelitian mendapati PLR pasien demensia berbeda dengan pasien normal. Penelitian ini bertujuan merancang algoritma computer vision yang dapat mendeteksi pupil secara akurat, menampilkan respons pupil terhadap cahaya dalam bentuk grafik dan PLR pada sebuah aplikasi desktop, yang mengendalikan goggles berisi rangkaian kamera, pencahayaan, dan sensor jarak VL53L0X. Rekaman diproses dengan Local Binary Pattern (LBP) dan deteksi kontur untuk mendeteksi pupil. Data pengukuran diproses dan disimpan pada basis data lokal dan aplikasi web, sehingga tenaga medis dapat menentukan ada atau tidaknya gejala demensia pada pasien lansia. Tingkat ketelitian algoritma pengukuran pupil sebesar 73,33% yang didapatkan dari 30 kali pengujian. Kata kunci: computer vision, demensia, deteksi dini, pupillary light response, pupilometri  ABSTRACT Pupillometry is a method of measuring the pupil’s response towards stimulus. Pupil response to light is observed through pupillary light response (PLR). Research found that PLR values of patients suffering from dementia differ from that of normal patients. This study implements a computer vision algorithm that accurately detects the pupil, calculates, and shows its response towards light in graphs and PLR values on a desktop application which controls goggles that contain a camera, lighting setup, and the VL53L0X distance sensor. Video is processed using Local Binary Pattern (LBP) and contour detection to detect the pupil. Results are processed and saved in the local and web database, so experts can determine the presence of dementia symptoms in the elderly patient. The accuracy of the pupil detection algorithm is 73,33%, as obtained from 30 tests. Keywords: computer vision, dementia, early detection, pupillary light response, pupillometry

The management and accountability in the activities of regional autonomy and local finance are regulated by Government Regulation Number 1 of 2022, which states that financial management at the regional level starts from the initial planning to the supervision of the regional finances. The success of performance improvement is deemed effective when it aligns with government regulations. Kabupaten Tegal is one of the regions whose capacity to self-finance governance, development, and public services is still reliant on transfers from the central and provincial governments. This is evident in the low Regional Financial Independence Ratio, which stands at 22.10%. Kabupaten Tegal has yet to fully implement fiscal decentralization, as indicated by the fluctuating fiscal decentralization ratio, which falls into the low category. To provide better public services, Kabupaten Tegal must optimize its Local Own-Source Revenue (PAD) potential. [ABSTRACT FROM AUTHOR]

Ekspresi wajah menjadi bahasa yang universal. Bahkan perubahan ekspresi wajah dapat membantu pengambilan keputusan. Pada tahun 1972, Paul Ekman mengklasifikasikan emosi dasar manusia ke dalam enam jenis: senang, sedih, terkejut, marah, takut, dan jijik. Kemudian Ekman dan Wallace Friesen mengembangkan sebuah alat untuk mengukur pergerakan pada wajah yang disebut Facial Action Coding System (FACS). FACS menentukan ekspresi wajah berdasarkan pergerakan otot wajah, yang diistilahkan Action Unit (AU). Penelitian ini bertujuan untuk mengetahui emosi tertarik yang dialami seseorang berdasarkan AU yang telah ditentukan oleh Paul Ekman dengan cara membandingkan dua buah citra, yaitu citra wajah tanpa ekspresi dan citra wajah berekspresi. Hasil penelitian ini memperoleh sebuah aplikasi yang mampu mengidentifikasi emosi tertarik dengan akurasi sebesar 80%, True Positive Rate 80%, dan True Negative Rate 80%. Dengan adanya penelitian ini diharapkan dapat diketahui karakteristik action unit yang membentuk emosi tertarik, juga memberikan masukan bagi proses evaluasi belajar mengajar mata kuliah pemrograman.

This study aims to propose and implement a Fall Detection System (FDS) for the Elderly Activity Daily Living (ADL) based on the modified two-dimensional (2D) Light Detection and Ranging (LiDAR) method, inspired by slice feature on three-dimensional (3D) LiDAR data. LiDAR-based FDS is an ambient-based FDS that utilizes infrared laser light. This work is an improvement from our previous proposed method, namely FDS based on 2D LiDAR data. Previous work presents a research gap analysis of falling and not falling classes based only on footprint-scanned data. Therefore, detecting other essential fall activities is impossible, such as falling forward, backward, sideward, and others. The main points carried out in this paper are the design and realization of modified 2D LiDAR, primary dataset collection, an accuracy test using K-Nearest Neighbors (K-NN), Random Forest (RF), as well as Support Vector Machine (SVM) algorithms. This work is comparable to the previous work that uses Infrared Array-based FDS. The test results in this paper prove that the modified 2D LiDAR has better performance than the previous study, especially in terms of sensitivity and selectivity performance parameters. The test results show that FDS achieves optimal performance using the RF algorithm with 98.89% sensitivity and 99.38% selectivity. This percentage value is superior to the previous study, with 98% sensitivity and 93% selectivity. In addition, the results show that the modified 2D LiDAR is feasible to be used as an ambientbased FDS solution. [ABSTRACT FROM AUTHOR]

Malaria is a communicable disease caused by a plasmodium parasite and transmitted among human by Anopheles mosquitoes. Late medication of this disease can cause a death of patients. Indonesia has many endemic areas with a high volume of patients diagnosed by Malaria. Currently, this incidences data are stored in Microsoft Excel files. We need to build a data warehouse to easily manage these data. Here, we create ontology of Malaria’s incidence data to figure out the important information in Malaria data warehouse that we want to build. We identify entities, classes, subclasses, and relationships between these entities. We employed Protégé to build and visualize the ontology of Malaria’ incidence data.

The 21st-century skills needed to face the speed of understanding technology. Such as critical thinking in computer vision and robotics literacy, any student is hampered by the programming that is considered complicated. This study aims at the improvement of student embedded system programming competency with computer vision and mobile robotics integration approach. This method is proposed to attract the students to learn about embedded system programming by delivering integration between computer vision and robotics using the P-D controller since both of the fields are closely related. In this paper, the researcher described computer vision programming to get the data of captured images through the camera stream and then delivered the data into an embedded system to make the decision of robot movement. The output of this study is the improvement of a student's ability to make an application to integrate a sensor system using a camera and the mobile robot running follow the line. The result of the test shows that the integration method between computer vision and robotics can improve the student's programming comprehension by 40%. Based on the Feasibility test survey, it can be interpreted that from the whole assessment after being converted to qualitative data, all aspects of the learning stages of programming application tested with the integration of computer vision and robotics fall into the very feasible category for used with a percentage of feasibility by 77.44%. [ABSTRACT FROM AUTHOR]