Bimodal Extended Kalman Filter-Based Pedestrian Trajectory Prediction.

We propose a pedestrian trajectory prediction algorithm based on the bimodal extended Kalman filter. With this filter, we are able to make full use of the dual-state nature of the pedestrian movement, i.e., the pedestrian is either moving or remains stationary. We apply the dual-mode probability model to describe the state of the pedestrian. Based on this model, we construct the proposed bimodal extended Kalman filter to estimate pedestrian state distribution. The filter obtains the state distribution for each pedestrian in the scene, respectively, and use that state distribution to predict the future trajectories of all the people in the scene. This prediction method estimates the prior probability of each parameter of the model through the dataset and updates the individual posterior probability of the pedestrian state through the bimodal extended Kalman filter. Our model can predict the trajectory of every individual, by taking the social interaction of pedestrians as well as the surrounding physical obstacles into account, with less than fifty model parameters being used, while with the limited parameter, our model could be nearly accurate as other deep learning models and still be comprehensible for model users.

Planning and Design of a Full-Outer-Air-Intake Natural Air-Conditioning System for Medical Negative Pressure Isolation Wards.

In the beginning of 2020, the coronavirus (COVID-19) pandemic started to spread globally, causing panic to the lives of people around the world; many countries executed lockdown of cities or even total lockdown of the entire countries. The coronavirus disease (COVID-19) is transmitted via air droplets. In medical environments that use traditional hermetic ventilation systems, medical personnel who come in contact with patients are more susceptible to infection compared to regular staff; therefore, the air flow and air quality of hermetic negative pressure isolation wards are highly critical. For this purpose, the study proposes a full-outer-air-intake natural air-conditioning system for negative pressure isolation wards. This innovative system draws in large amounts of fresh external air to greatly improve the air exchange rate in wards; negative pressure environments can be implemented depending on requirements to solve the issue of nosocomial infections in traditional negative pressure isolation wards that draw air from within the hospital. This greatly reduces the probability of nosocomial infection and infection via air droplets; furthermore, the system's intake and exhaust paths are completely isolated, solving the issue of air cross-contamination. Based on the results from the experiment site, this innovative system was designed and implemented based on the guidelines of hospital facilities and achieved air exchange per hour in excess of 12 times/hour, reaching a maximum of 54.5 times/hour. Indoor CO2 concentration was 576 ppm, negative pressure was -14 Pa, indoor temperature was 23.3°C, indoor humidity was 54.1%, and sensible heat exchange efficiency (ηs) was 105.88% which effectively reduced ventilation load. Therefore, this innovative full-outer-air-intake natural air-conditioning system can provide medical staff and patients with a safe and healthy environment that prevents cross-infection.

Design and Implementation of a Low-Energy-Consumption Air-Conditioning Control System for Smart Vehicle.

About 7% of people's daily time is spent in taking vehicles between office and home. Besides, with the improvement of the living standard in today's society, people's requirements for a comfortable environment inside the car are constantly increasing and this must rely on an effective vehicle air conditioner to maintain the comfort of the cabin environment. In general, a vehicle air conditioner uses the air-mixing mode to regulate the temperature control system. In this mode of operation, the compressor needs to work continuously, which is extremely energy consuming. The vehicle's air conditioner is greatly affected by the inner and outer heat load, which are generated therein. Furthermore, the heat load is instantly changeable. Therefore, only when the controller can adapt to the feature of heat load, then we can find the optimal control method, thus enabling the vehicle's air conditioner to interact with the actual heat load to supply the balanced cooling capacity and, as a result, create the most comfortable environment inside the cabin with minimum energy consumption. For this purpose, we bring up in this paper a low-energy-consumption smart vehicle air-conditioning control system to detect total heat load, which can change the vehicle's air-conditioning capacity mode to maintain the average temperature at 25.2°C∼26.2°C and the average humidity at 46.6%∼54.4% in the cabin. When the inner heat load is stable, the rest times of the compressor can reach 16∼23 times per hour, which attains a rate of fuel saving around 21%∼28%. With the proposed architecture, the purpose of the low-energy-consumption vehicle air-conditioning system can be achieved, which, at the same time, creates a comfortable environment inside the cabin.

Planning and Design of Low-Power-Consuming Full-Outer-Air-Intake Natural Air-Conditioning System.

A person stays indoors for about 85%∼90% time of his lifetime, and the need for a comfortable indoor environment is getting higher; thus, the air-conditioning dependency becomes intense too. Nowadays, residents focus on both the comfortable living environment and indoor air quality. A closed environment will become hazardous because of carbon dioxide released during respiration and toxic organic solvent vapor released from interior decoration. In order to improve the indoor air quality (IAQ), we must allow outer fresh air into the indoor space and release the dirty air out. But while taking in fresh air, the heat and factory/vehicle exhaust are also introduced. Indoor CO2, HCHO, and VOCs and outer dirty gas threaten human health badly. To solve this problem, we bring up an innovative low-power-consuming full-outer-air-intake natural air-conditioning system that completely separates intake and exhaust air, which is a solution for cross-contamination and makes mass/energy exchange by means of air and water. Design airflow exceeds 300∼500 CFM, steam evaporation mass rate reaches 3.13∼3.88 kg/hr, and heat exchange capacity becomes 1,855∼2,300 kcal/hr. The sensible heat effectiveness is 71%∼112%, and EER exceeds 14.05∼17.42 kcal/W·h. In addition, the system under design can be of positive or negative pressure status according to the user's or work's requirement. It creates a comfortable and healthy living environment by supplying clean and fresh outer ambient air with low power consumption.

Body posture recognition and turning recording system for the care of bed bound patients.

This paper proposes body posture recognition and turning recording system for assisting the care of bed bound patients in nursing homes. The system continuously detects the patient's body posture and records the length of time for each body posture. If the patient remains in the same body posture long enough to develop pressure ulcers, the system notifies caregivers to change the patient's body posture. The objective of recording is to provide the log of body turning for querying of patients' family members. In order to accurately detect patient's body posture, we developed a novel pressure sensing pad which contains force sensing resistor sensors. Based on the proposed pressure sensing pad, we developed a bed posture recognition module which includes a bed posture recognition algorithm. The algorithm is based on fuzzy theory. The body posture recognition algorithm can detect the patient's bed posture whether it is right lateral decubitus, left lateral decubitus, or supine. The detected information of patient's body posture can be then transmitted to the server of healthcare center by the communication module to perform the functions of recording and notification. Experimental results showed that the average posture recognition accuracy for our proposed module is 92%.

A smart phone-based pocket fall accident detection, positioning, and rescue system.

We propose in this paper a novel algorithm as well as architecture for the fall accident detection and corresponding wide area rescue system based on a smart phone and the third generation (3G) networks. To realize the fall detection algorithm, the angles acquired by the electronic compass (ecompass) and the waveform sequence of the triaxial accelerometer on the smart phone are used as the system inputs. The acquired signals are then used to generate an ordered feature sequence and then examined in a sequential manner by the proposed cascade classifier for recognition purpose. Once the corresponding feature is verified by the classifier at current state, it can proceed to next state; otherwise, the system will reset to the initial state and wait for the appearance of another feature sequence. Once a fall accident event is detected, the user's position can be acquired by the global positioning system (GPS) or the assisted GPS, and sent to the rescue center via the 3G communication network so that the user can get medical help immediately. With the proposed cascaded classification architecture, the computational burden and power consumption issue on the smart phone system can be alleviated. Moreover, as we will see in the experiment that a distinguished fall accident detection accuracy up to 92% on the sensitivity and 99.75% on the specificity can be obtained when a set of 450 test actions in nine different kinds of activities are estimated by using the proposed cascaded classifier, which justifies the superiority of the proposed algorithm.

A three-step approach with adaptive additive magnitude selection for the sharpening of images.

Aimed to find the additive magnitude automatically and adaptively, we propose a three-step and model-based approach for the sharpening of images in this paper. In the first pass, a Grey prediction model is applied to find a global maximal additive magnitude so that the condition of oversharpening in images to be sharpened can be avoided. During the second pass, edge pixels are picked out with our previously proposed edge detection mechanism. In this pass, a low-pass filter is also applied so that isolated pixels will not be regarded as around an edge. In the final pass, those pixels detected as around an edge are adjusted adaptively based on the local statistics, and those nonedge pixels are kept unaltered. Extensive experiments on natural images as well as medical images with subjective and objective evaluations will be given to demonstrate the usefulness of the proposed approach.

An efficient and self-adapted approach to the sharpening of color images.

An efficient approach to the sharpening of color images is proposed in this paper. For this, the image to be sharpened is first transformed to the HSV color model, and then only the channel of Value will be used for the process of sharpening while the other channels are left unchanged. We then apply a proposed edge detector and low-pass filter to the channel of Value to pick out pixels around boundaries. After that, those pixels detected as around edges or boundaries are adjusted so that the boundary can be sharpened, and those nonedge pixels are kept unaltered. The increment or decrement magnitude that is to be added to those edge pixels is determined in an adaptive manner based on global statistics of the image and local statistics of the pixel to be sharpened. With the proposed approach, the discontinuities can be highlighted while most of the original information contained in the image can be retained. Finally, the adjusted channel of Value and that of Hue and Saturation will be integrated to get the sharpened color image. Extensive experiments on natural images will be given in this paper to highlight the effectiveness and efficiency of the proposed approach.