Face is a strong feature for person detection and identification. Face
images are, however, degraded in bad illumination conditions such as
back-lighting. We have developed an illumination normalization method
that coverts faces in any illumination conditions to fairly consistent
appearance, thereby making models for face detection and
identification be very robust. This conversion is done by a
GA-optimized fuzzy inference and runs in real-time.
[References]
B.S.B. Dewantara and J. Miura,
"OptiFuzz: A Robust Illumination Invariant Face Recognition System
and Its Implementation", Machine Vision and Applications, 2016．
2. Human Facing Direction Detection For Identifying Attention
The human head and face are the most common parts of the human body used in computer vision applications such as detecting the presence of a person, identifying and verifying a person, and indicating one's attention. Head orientation can be used to estimate an attentional awareness during an interaction process so that one may expect an appropriate response that is in-line with the degree of attention. We develop a novel descriptor that combines more features in order to discriminate various head orientations.
3. A Socially Acceptable Guide Robot Navigation Control
Navigating a robot in a human or social environment is a difficult task since the robot should understand the human-human interaction rules. This means that the robot must navigate gently, smoothly and safely among human surrounding. We have developed a modified Social Force Model (SFM) for controlling the robot movements by considering adaptive inner parameters adjustment as the representation of human feeling and psychological aspects.
[References]
Bima Sena Bayu Dewantara and Jun Miura, "Generation of a Socially Aware Behavior of a Guide Robot Using Reinforcement Learning", IEEE International Electronic Symposium (IES), 2016.
4. Human Detection From Body Shape
In human-machine interaction, the existence of human is very important to be considered. Many existing methods tried to detect human based on photometrical information which is frequently disturbed by external disturbance, such as lighting conditions. We have developed an indoor human detection system by utilizing the shape of human's depth-image. Histogram of Oriented Gradient (HOG) and Support Vector Machine (SVM) are used to realize our system.
[References]
2. Human Facing Direction Detection For Identifying Attention
The human head and face are the most common parts of the human body used in computer vision applications such as detecting the presence of a person, identifying and verifying a person, and indicating one's attention. Head orientation can be used to estimate an attentional awareness during an interaction process so that one may expect an appropriate response that is in-line with the degree of attention. We develop a novel descriptor that combines more features in order to discriminate various head orientations.
[References]
B.S.B. Dewantara and J. Miura, "Estimating Head Orientation Using a Combination of Multiple Cues", IEICE Transaction on Information and Systems, 2016.3. A Socially Acceptable Guide Robot Navigation Control
Navigating a robot in a human or social environment is a difficult task since the robot should understand the human-human interaction rules. This means that the robot must navigate gently, smoothly and safely among human surrounding. We have developed a modified Social Force Model (SFM) for controlling the robot movements by considering adaptive inner parameters adjustment as the representation of human feeling and psychological aspects.
Bima Sena Bayu Dewantara and Jun Miura, "Generation of a Socially Aware Behavior of a Guide Robot Using Reinforcement Learning", IEEE International Electronic Symposium (IES), 2016.
4. Human Detection From Body Shape
In human-machine interaction, the existence of human is very important to be considered. Many existing methods tried to detect human based on photometrical information which is frequently disturbed by external disturbance, such as lighting conditions. We have developed an indoor human detection system by utilizing the shape of human's depth-image. Histogram of Oriented Gradient (HOG) and Support Vector Machine (SVM) are used to realize our system.
Bima Sena Bayu Dewantara, Fernando Ardilla, and Ardiansyah At Thoriqy, "Implementation of Depth-HOG based Human Upper Body Detection On A Mini PC Using A Low Cost Stereo Camera", International Conference of Artificial Intelligence and Information Technology (ICAIIT), 2019.
5. Trash Detection System
Whether aware or not, the human is the biggest trash contributor on earth. The rapid growth of human population making the human needs growing fast. Surely one day, from the human needs, will produce a rapid growth of waste when the growth is not controlled with the trash handling habit. We conducted a work to create a social education trash bin robot by visually detecting and classifying trash to organic and non-organic waste. The robot will autonomously travel around in public facilities to scanning for trash.
[References]
Irfan Salimi, Bima Sena Bayu Dewantara, and Iwan Kurnianto Wibowo, "Visual-based trash detection and classification system for smart trash bin robot", 2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing (IES-KCIC), 2018.
6. IoT-Based Human-Robot Communication
We have developed communication system between robot and operator based on IoT by utilizing telegramBot. By using telegramBot, operator can monitor and supervise robot works and will receive notifications when a robot has several conditions such as low battery, the trash is full and the robot should be stopped working because of system on demand.
[References]
Kisron, Bima Sena Bayu Dewantara, and Fernando Ardilla, "Early Warning and IoT-based Reporting System for Mobile Trash Bin Robot Application", 2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing (IES-KCIC), 2018.
5. Trash Detection System
Whether aware or not, the human is the biggest trash contributor on earth. The rapid growth of human population making the human needs growing fast. Surely one day, from the human needs, will produce a rapid growth of waste when the growth is not controlled with the trash handling habit. We conducted a work to create a social education trash bin robot by visually detecting and classifying trash to organic and non-organic waste. The robot will autonomously travel around in public facilities to scanning for trash.
Irfan Salimi, Bima Sena Bayu Dewantara, and Iwan Kurnianto Wibowo, "Visual-based trash detection and classification system for smart trash bin robot", 2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing (IES-KCIC), 2018.
6. IoT-Based Human-Robot Communication
We have developed communication system between robot and operator based on IoT by utilizing telegramBot. By using telegramBot, operator can monitor and supervise robot works and will receive notifications when a robot has several conditions such as low battery, the trash is full and the robot should be stopped working because of system on demand.
Kisron, Bima Sena Bayu Dewantara, and Fernando Ardilla, "Early Warning and IoT-based Reporting System for Mobile Trash Bin Robot Application", 2018 International Electronics Symposium on Knowledge Creation and Intelligent Computing (IES-KCIC), 2018.
