The paper presents an innovative method to artfully present multiple human and nonhuman viewpoints on a user’s current viewpoint. The method is built on a digital platform for capturing a large number of human and nonhuman viewpoints in the world, and enabling us to collectively share the viewpoints. Inspired by the art, the presentation method enables people to watch the world in a novel way.
In precision livestock farming, lameness detection in cattle is particularly important for breeding management. The accurate detection of lameness is crucial for delivering effective and economical treatment and for preventing future diseases. The noticeable sign of lameness is that their speed of walking, arching their backs and drop their heads during walking. Here, we emphasis on lameness of dairy cattle by implementing the intelligent visual perception system on the laneways after milking process. Employing a deep learning technique of Mask-RCNN for cattle region detection and identification. The novelty of this work noticeably implies that deep learning instance segmentation could be effectively employed as a cattle region extraction from complex background prior to using identification and tracking.
This paper presents a sensor-based data acquisition glove for gesture recognition in Japanese Sign Language (JSL), which uses five flex sensors and an inertial measurement unit (IMU) to detect finger flexion and hand motion information. The detected data is sent from the Arduino Micro to a computer. We collected data from the "A" to "Ta" lines of the Japanese (kana) syllabary and using four different machine learning algorithms: Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF), and K-Nearest Neighbor (KNN) to recognize them. RF and KNN have the highest average accuracy, reaching 99.75%. Also, SVM and DT had an average accuracy of 99% and 94.25% respectively. The experimental results show that the proposed system has great potential for gesture recognition in Japanese Sign Language.
Cow identification has become important in recent years due to outbreaks of diseases such as bovine spongiform encephalopathy. Conventional identification methods available are not efficient, affordable, non-invasive, and cost- effective. Among them, some methods are based on biological markers, such as muzzle point matching and facial recognition. Facial images are the most common biometric characteristics used by humans to identify individuals, and they have received much attention. In this study, we used RGB camera to identify individual cow by their faces and confirmed the effectiveness of this method.
The economic depression accompanied by COVID-19 pandemic has also had some adverse effects on people's mental health. 4 in 10 adults have symptoms of anxiety and depression, and these symptoms may continue to increase. In particular, the reported rate of mental health disorders among adolescents is as high as 56.2%. Recently, pet ownership has become widely accepted as a means of improving mental health. However, due to various reasons, the separation of owners and pets has also increased significantly. Therefore, we have developed a portable simulated pet "KEDAMA", which can be used in place of real pets, and can be customized with fur and sound. The price is relatively cheap and can be ordered easily. We conducted a statistical analysis on 20 users, and the results show that the "KEDAMA" has a positive effect on users' mental health.
In this research, two experiments were performed to develop a noncontact measurement method for flow velocity in nasal breathing. On the basis of the results of our preliminary study, we first compared the instantaneous flow velocity of nasal breathing and the air temperature of inhalation and exhalation. Next, we compared the air temperature and the nasal regions on facial thermal images. Results showed that the correlation between flow velocity, air temperature, and the nasal region can be improved using a smoothing filter and our modified normalization procedure.
In this study, we propose color inspection tables that detect a degree of color weakness. Hereafter, a creating method of color inspection tables will be described. Firstly, we find two colors that cannot be discerned by the color-defective vision and can be discerned by the normal vision. Secondly, we create a color inspection table consisting of 8 patches using the confused two colors. We make 7 patches with the first color and 1 patch with the second color. Here, we should consider the display order of the color inspection tables so as not to affect the results by the recency effect. The collaborators in the experiment select one different color patch from the color inspection table displayed on the monitor. Three color inspection tables are used to detect the degree of color weakness in 10 experiment collaborators. From the experimental results, it is confirmed that the color inspection tables can measure the degree of color-defective vision.
Data conversion of the current system is essential for the commercialization of autonomous buses. The biggest advantage of autonomous buses is that they can follow flexible routes according to traffic demand. If a program automatically finds the optimal route depending on the time of day or traffic demand, then it will be very helpful for the commercialization of autonomous buses. Because it is most important to gather data such as the locations of bus stops, this paper proposes a method of creating a platform based on the bus stop data provided by Daegu City. Currently, Daegu City provides all locations and demand for bus stops. Visual Studio was used to classify the data according to Daegu's districts, and a platform to find the optimal route was created using the modified Dijkstra algorithm, which made it possible to simulate flexibly according to changes in data.
In this paper, we describe a new air-conditioning system that takes human behavior into account. We introduce a method of predicting the indoor comfort level distribution by combining computational fluid dynamics (CFD) and Internet of Things (IoT), and a concept of air-conditioning guidance that encourages human behavior by using this method. In particular, we discuss the method of optimizing parameter for the CFD analysis used in the system, which ensures acceptable computational speed and accuracy in the proposed guidance system, by comparing the results of experiments using a small room model.
In recent decades, annual number of extremely hot days with a maximum temperature of above 35°C is tend to increase worldwide. Such hot environments are associated with increased risk of heat stroke. Therefore, an accurate heat stress assessment method is required to estimate the heat stroke risk and to find effective countermeasures for heat stroke. However, the conventional heat stress indices, such as wet-bulb globe temperature (WBGT), cannot straightforwardly apply extremely hot environments that exceed body temperature because of little experimental evidence to support the applicability the conventional standards to such environments. In this paper, we propose a novel heat stress assessment method using the group mean heart rate response measured by a wearable biosensor network. Our method can provide real-world evidence of the heat stress that workers are exposed. The heat stress index of our method can be derived from real-world data. We will discuss the usefulness of this method based on the real-world data measured in workplace environments.
This study developed a comprehensive screening test system for attention disorders in right hemisphere damage, which works with a tablet computer and stylus pen. This system can detect more symptoms with the same ease as a paper-based test.
As people lead safer and healthier work and lives, aspirations of well-being and health-rated quality of life. Wellbeing exists in two dimensions: Subjective well-being (SWB) is based on basic drive needs, psychological well-being (PWB) is based on assumptions about Ikigai. This study investigated whether the difference of well-being in varied work. The results suggest that for the meaning of working contents, the dimension of well-being may be benefited.
The authors proposed a new surgical instrument that can cut a target region of the liver from the surface into a cylindrical shape and remove a tumor with less loss of liver tissue. The instrument has a unique mechanism that drives the four blades three-dimensionally, alternating between cutting and coagulation. In order to confirm the effectiveness of the proposed mechanism, a desktop experimental prototype was manufactured. Through an experiment on a pig liver using a prototype, it was shown that the target portion can be excised into a cylindrical shape as expected by the proposed mechanism.
This paper describes a soft robotic hand for people with severe disabilities. We used soft material, thermoplastic polyurethane (TPU), to create the robotic hand. Using the TPU hand with a robotic arm, we experimentally performed to grasp and lift object. 50 kinds of daily needs were tested as grasping object in the experiment. From the experimental results, it is found that the success rates of grasping rectangular-, spheral/circular-, and cylindrical-shaped objects are more than 80 %, and that of grasping indeterminate-shaped objects is less than 77 %.
Unsuitable foot clearance and step length lead falling accidents; thus, these gait parameters should be monitored for fall prevention. According to previous study, users prefer wrist as a position to wear wearable device for daily gait monitoring. Therefore, we have been developing gait monitoring method using wearable device on a wrist. Our previous method classifies gaits with different foot clearance or step length by machine learning technique using 3-axis acceleration data obtained from wearable device on a wrist. The results of our previous study showed that this method could classify gaits for each step. However, accuracy of our previous method was insufficient because individual difference of arm acceleration due to stature or gait speed were not considered. The objective of this study is to propose the gait classification method based on individual difference. The proposed method classifies three gaits with different foot clearance and step length by machine learning technique using combination of arm acceleration, stature, and gait speed. The proposed method was tested for 1800 steps performed from ten participants. The proposed method could classify three gaits with accuracy greater than 0.9. These results indicate possibility that the proposed method can be used for gait monitoring using wearable device.
Mental fatigue is the cognitive state in which an individual feels tired, and mentally exhausted. Such phenomenon has been prevalently present in workspaces that reinforce prolonged work hours that result in overtime work by employees, hampering their productivity and moreover, their mental state and health. Mental fatigue has been linked with endangering mental behaviors such as anger, high levels of anxiety, and depression. Mental fatigue is commonly measured through indirect measurements. The focus of the current research aims to identify variables that are closely connected to monitoring an individual’s mental state. This research was conducted with the aim of utilizing the Mind wave, a hardware that utilizes one dry electrode to collect brain signals from the user and identify and predict mental fatigue thresholds through the use of deep neural networks. The proposed deep learning model was utilized to identify whether an individual is currently at a normal state or reaching mental fatigue thresholds through binary classification. Results obtained have shown that the proposed model can identify mental fatigue thresholds efficiently and perform well when compared to baseline methods.
Despite the declining birthrate and aging population, the labor shortage in the long-term care industry has not been improved due to the heavy burden on long-term care workers. Against this background, various transfer assistance robots have been proposed. We are currently developing a robot that can assist in transferring a patient from a nursing bed to a wheelchair. The transfer assistance robot needs to be created taking into account the various body size differences of the caregivers. This paper describes the dimensions and mechanism of the nursing robot according to the difference in body size. Also, we are developing a nursing care robot that can be transferred comfortably by reducing the physical and mental burden on the care recipient.
A breast cancer is known as one of the most dangerous disease of the death causes among aged 40-55 women. We need to develop a computer aided diagnosis system for breast cancer classification. In the previous study, the random forest of the ensemble learning method was reported to be one of promising classifiers for classifying breast cancers using a Wisconsin Diagnostic Breast Cancer(WDBC) dataset. Furthermore, we found the effectiveness of the random forest with recursive feature elimination(RFE) for breast cancer classification, compared to another ensemble learning methods, XGBoost and LightGBM(LGBM). The RFE is known as a feature selection method. In our experiments, however, there was no comparison with the feature extraction of principal component analysis(PCA). This paper presents the evaluation of feature extraction methods with ensemble learning for breast cancer classification on the WDBC dataset.
This paper presents a methodology for automated data entry of salary payslips from document images. The challenging problems are 1) the payslips vary from one company to another, and 2) the appeared wording terms are different but similar meaning terms. The proposed methodology is the essential preprocess by using image processing and regular expression setting before an optical character recognition or OCR. The post-process for number validation must be considered by checking the financial formula.
With the spread of the coronavirus infection in Japan, the suicide rate has increased. So, we thought that we want to take care of people’s mental health by using music. In previous studies, they showed that music included 1/f frequency influence the state of mind relaxing. On the other hand, there are reports that the natural sounds induce the state of relaxing. In this study, we investigated whether the reminiscence of scenes by natural sounds can work on the central nervous system and bring about a relaxing effect, and whether a synergistic effect can be observed by working together with the autonomic nervous system. Headphones and a questionnaire were also used. Twelve young male and female were recruited as subjects. The results showed that most of the subjects had a good impression of music, nature sounds and their mixtures. The results for music and nature sounds were mixed, but most of the subjects had a good impression of the blended sounds. The effect on relaxation, as measured by the questionnaire, was better after listening to the music than before in all conditions. In all conditions, there was a greater improvement in relaxation after listening to the music than before, and the degree of improvement was as follows: mixture > nature sounds > music. Although there was no statistical superiority, it was possible to show this trend. This suggests that there may be a synergistic effect of mixing music, which affects the autonomic nervous system, with nature sounds, which affect the central nervous system.
