Elmer Dadios
De La Salle University

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

Deep hypersphere embedding for real-time face recognition Ryann Alimuin; Elmer Dadios; Jonathan Dayao; Shearyl Arenas
TELKOMNIKA (Telecommunication Computing Electronics and Control) Vol 18, No 3: June 2020
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.12928/telkomnika.v18i3.14787

Abstract

With the advancement of human-computer interaction capabilities of robots, computer vision surveillance systems involving security yields a large impact in the research industry by helping in digitalization of certain security processes. Recognizing a face in the computer vision involves identification and classification of which faces belongs to the same person by means of comparing face embedding vectors. In an organization that has a large and diverse labelled dataset on a large number of epoch, oftentimes, creates a training difficulties involving incompatibility in different versions of face embedding that leads to poor face recognition accuracy. In this paper, we will design and implement robotic vision security surveillance system incorporating hybrid combination of MTCNN for face detection, and FaceNet as the unified embedding for face recognition and clustering.
Alignment control using visual servoing and mobilenet single-shot multi-box detection (SSD): a review Jayson Rogelio; Elmer Dadios; Argel Bandala; Ryan Rhay Vicerra; Edwin Sybingco
International Journal of Advances in Intelligent Informatics Vol 8, No 1 (2022): March 2022
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v8i1.819

Abstract

The concept is highly critical for robotic technologies that rely on visual feedback. In this context, robot systems tend to be unresponsive due to reliance on pre-programmed trajectory and path, meaning the occurrence of a change in the environment or the absence of an object. This review paper aims to provide comprehensive studies on the recent application of visual servoing and DNN. PBVS and Mobilenet-SSD were chosen algorithms for alignment control of the film handler mechanism of the portable x-ray system. It also discussed the theoretical framework features extraction and description, visual servoing, and Mobilenet-SSD. Likewise, the latest applications of visual servoing and DNN was summarized, including the comparison of Mobilenet-SSD with other sophisticated models. As a result of a previous study presented, visual servoing and MobileNet-SSD provide reliable tools and models for manipulating robotics systems, including where occlusion is present. Furthermore, effective alignment control relies significantly on visual servoing and deep neural reliability, shaped by different parameters such as the type of visual servoing, feature extraction and description, and DNNs used to construct a robust state estimator. Therefore, visual servoing and MobileNet-SSD are parameterized concepts that require enhanced optimization to achieve a specific purpose with distinct tools.
Gabor-enhanced histogram of oriented gradients for human presence detection applied in aerial monitoring Anton Louise Pernez De Ocampo; Argel Bandala; Elmer Dadios
International Journal of Advances in Intelligent Informatics Vol 6, No 3 (2020): November 2020
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v6i3.514

Abstract

In UAV-based human detection, the extraction and selection of the feature vector are one of the critical tasks to ensure the optimal performance of the detection system. Although UAV cameras capture high-resolution images, human figures' relative size renders persons at very low resolution and contrast. Feature descriptors that can adequately discriminate between local symmetrical patterns in a low-contrast image may improve a human figures' detection in vegetative environments. Such a descriptor is proposed and presented in this paper. Initially, the acquired images are fed to a digital processor in a ground station where the human detection algorithm is performed. Part of the human detection algorithm is the GeHOG feature extraction, where a bank of Gabor filters is used to generate textured images from the original. The local energy for each cell of the Gabor images is calculated to identify the dominant orientations. The bins of conventional HOG are enhanced based on the dominant orientation index and the accumulated local energy in Gabor images. To measure the performance of the proposed features, Gabor-enhanced HOG (GeHOG) and other two recent improvements to HOG, Histogram of Edge Oriented Gradients (HEOG) and Improved HOG (ImHOG), are used for human detection on INRIA dataset and a custom dataset of farmers working in fields captured via unmanned aerial vehicle. The proposed feature descriptor significantly improved human detection and performed better than recent improvements in conventional HOG. Using GeHOG improved the precision of human detection to 98.23% in the INRIA dataset. The proposed feature can significantly improve human detection applied in surveillance systems, especially in vegetative environments.
Radial greed algorithm with rectified chromaticity for anchorless region proposal applied in aerial surveillance Anton Louise Pernez De Ocampo; Elmer Dadios
International Journal of Advances in Intelligent Informatics Vol 5, No 3 (2019): November 2019
Publisher : Universitas Ahmad Dahlan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26555/ijain.v5i3.426

Abstract

In aerial images, human figures are often rendered at low resolution and in relatively small sizes compared to other objects in the scene, or resemble likelihood to other non-human objects. The localization of trust regions for possible containment of the human figure becomes difficult and computationally exhaustive. The objective of this work is to develop an anchorless region proposal which can emphasize potential persons from other objects and the vegetative background in aerial images. Samples are taken from different angles, altitudes and environmental factors such as illumination. The original image is rendered in rectified color space to create a pseudo-segmented version where objects of close chromaticity are combined. The geometric features of segments formed are then calculated and subjected to Radial-Greed Algorithm where segments resembling human figures are selected as the proposed regions for classification. The proposed method achieved 96.76% less computational cost against brute sliding window method and hit rate of 95.96%. In addition, the proposed method achieved 98.32 % confidence level that it can hit target proposals at least 92% every time.