Within the past 2 decades, several unsatisfactory performance of construction safety has been recorded in Hong Kong. Surprisingly, it has been found 1700 injuries with 6 fatal cases in the civil engineering and construction industry in the first half of 2018 only. Although there is a decreasing trend of construction safety accidents reflected by statistical figures, the inaccuracy of construction health and safety data still exists, namely the intentious cheating and negligence in treating the safety audit data. Such fake information not just hinders public to realise the true picture of construction accidents, but also causes significant economic loss to the society. In this regard, “Construction Safety Audit with Blockchain” (CSAB) under Smart Contract Platform (SCP) and hyperledeger, which is innovative to the entire construction industry, proposed to eliminate the human errors on safety auditing in construction industry. Composing of smart contract and hyperledger, this blockchain can allow the exchange of construction safety information between different interested parties and monitored by the platform. It will be also be a great honour for CityU to have this innovative platform of blockchain to contribute the construction industry, as well as becoming the leader of blockchain in local universities.
Regenerative braking system are used on vehicle, so it is believed that it can also be applied on electric or hybrid plane. Benefit from the fan wing design, the rotating fan can generate lift and drag at the same time which can be made at a good use for regenerating some electricity for emergency case.
Nowadays, it is more popular to purchase daily products or fresh raw food through delivery. It can be foreseen that there are still a large potential of delivery proven by the thriving of Amazon in America, SF Express in Mainland China and HKTV in Hong Kong. Moreover, the recent COVID-19 raises public awareness to hygiene. They prefer purchase online to reduce the chance to go outside to protect themselves. In order to optimize the efficiency and effectiveness of the delivery system, we propose a delivery system that could take advantage of the modern development in drones to transport goods, therefore achieving a high efficiency and low cost of delivery. With the rapid development of e-commerce, the market potential of delivery would continue to growth. With the use of drone, the industry could reduce the reliance on labor, especially in places like Hong Kong which the labor costs is very high, and so we could enjoy a cheaper goods without the needs of leaving home, open up more business opportunity for “remote” places. The delivery system would use drone to deliver the goods to the target customer. The system could be mostly controlled by computer with minimal human intervention.
Medical professionals are core to our society who bear responsibility for saving and mending lives. Often times, the key transition between life and death is medical surgery. Thus, when dealing with such consequential matters, medical practitioners must be equipped with practical skills and experience to operate seamlessly. Among all medical disciplines, this project retains a focus on intraocular eye surgery, which is one of the most common and important types of surgeries in the world. Nowadays, the likelihood of contracting the dry-eye disease has increased due to technology-driven lifestyles. This leads to more serious eye illnesses or even blindness. Since the eye is one of the most fragile organs of the human body, intraocular surgery requires high accuracy and precision. Despite this, relevant training platforms are not readily available, accessible nor affordable, especially in developing regions. Currently, Eyesi® Surgical is the most popular ocular surgery training platform for medical students and professionals. However, each unit costs about 1 million dollars. Even Hong Kong medical schools, such as CUHK only have a couple which is sufficient for training and practice for the future doctors of Hong Kong, let alone developing countries, according to a prominent CUHK medical professor. To tackle this, our team has produced an affordable and portable ocular microsurgery prototype that costs less than HKD7000, equipped with quality software modules and hardware components. In other words, a firm contender against professional eye surgery training technologies such as Eyesi® Surgical, but at a fraction of the cost and weight. Additionally, it is compatible with virtual reality and has its own mobile application supporting its PC client equivalent.
It is a smart sheet for musicians and busker. Just simply type in the song name, it will then google you the chord sheet and show it in the editor. Modify your music sheet as you desire, then go to the next page. This page include several main functions: – Detect the key of the song. – Separating the lyrics and chords by changing them into different colours. – Change Key, front size and the spacing of the song.
With technology advancement, smartphones nowadays are commonly integrated with high-resolution cameras, incorporating multi-lens features, precise AI support and optical zoom capabilities, which deliver quality photography experience. Given the great accessibility and reliable performance, more are eager to opt for a smartphone when it comes to photography. In order to provide a more accessible way to review the precious moments captured by our smartphones, we came up with an idea to upgrade the existing digital frames in several aspects for a more user-friendly experience. We found out that most digital frames currently available in the market are featured with built-in storage and syncing function, yet they are constrained to be placed on flat surface, with limited size and unsatisfying resolution. Therefore, we would like to put forth our idea to improve the design based on current features. Not only does our proposed digital frame feature in cloud services with Instagrm, Facebook etc, and remote control with smartphones, it is also capable of being hung on a whiteboard-like wall mount with inductive charging function and free-angle rotation. We believe our 20-inch design with 1080p HD display would provide a quality using experience and other features would be discussed in our proposal.
Shelfitor – your very own shelf monitor. When we visited the elderly centres in Hong Kong, we noticed incredible organization. All shelf racks were designated to one item, and elderly care-takers would come in once a week to monitor the stock. With numerous rooms, they would manually visit each room, monitor stock, then go organize stock and then rearrange it. I remember one of the elderly members telling me – the care-takers were once able to spend quality time with them, but now, because of the amount of workload they have, they no longer could. Mrs Ashley Choi , an elderly care worker in the centre, she told us that a worker was assigned to every floor for this activity and spent 20 man hours a week just doing this activity. 11 workers in any residential care home for elderly in HK, will be spending 220 man hours engaging in this activity, instead of spending quality time with the aging population of Hong Kong. Our tiles simply have to be placed on the rack of a shelf, and are in-built with technology that recognizes the weight of items placed on it, and stores any changes. If the weight of the shelf falls below a certain threshold, an alert will be shown on the dashboard, from which care-takers can easily detect shortage, and replenish accordingly.
Scoliosis is a medical condition where one’s spine curves sideways. As is commonly observed in children, scoliosis needs diagnosing early to avoid deterioration. Conventionally, doctors spend much time studying the back X-ray of the patients. Modern methods use deep learning to process the X-ray. However, the radiation given out by X-ray machines, which hurts children’s health, cannot be avoided by these methods. We present a new workflow to diagnose scoliosis by synthesizing the X-ray from the RGB-D images of the back of the patients using deep learning models, which can rid the whole process of radiation. We used anatomical landmarks to reveal shape of the spine. Firstly, landmark detection on RGB-D images using the High-Resolution Nets was conducted. Then we used the pix2pix model to synthesize X-ray from RGB-D images and landmarks. Detailed experiments were conducted to tune the models. Our synthetic images were clear and close to the ground truth. Only 520 training samples were available before the deadline. However, good performance on this small dataset could prove our idea. If more data is available, X-ray machines are no longer needed for diagnosis. Our project has a bright prospect.
This report describes a study to the usefulness of implementing and integrating machine learning and augmented reality towards dyslexia children education. Dyslexia is a reading and writing disorder, children who suffer from dyslexia require extra effort to understand each word, and it may lead to higher frustration in the learning process. The aim of this project is to encourage and help dyslexia children to learn Chinese characters in a self-directed way through gamification of training exercise and machine learning. There are three major components in our solution which applied augmented reality and machine learning. The augmented reality part is implemented in a scramble-puzzle game component; the machine learning part is implemented in a Chinese character recognition component through mobile camera input and a handwriting training exercise component. The study will then evaluate the result of the implementation of our solution. The performance evaluation got a satisfactory result to our application performance. The user evaluation indicates that machine learning and augmented reality are suitable for dyslexia educational purposes and somehow motivated the dyslexia children to learn.
Hong Kong’s rail system which offers connectivity throughout the city has a good performance record in terms of reaching destinations on time. However, rail systems are not foolproof and may suffer from disruptions, such as technical failures and accidents, leading to delays. At present, there exists no conclusive technological response platform for mitigating the impact of such mishaps and offering relevant, real-time recommendations for optimizing the handling of incidents. The thesis project explores a solution in the form of a recommendation engine which can simulate various scenarios and thereby propose effective responses to potential mishaps. The report therefore, acts as a platform for discussing the potential application, benefits and challenges of utilizing technology in transport systems and in particular, railway networks. The work discusses the available technological choices and various development approaches and compares them, also discussing relevant works of literature offering domain expertise, constraints faced during the project and remedial measures taken to tackle them. Future prospects of the project include automation of the entire rail network, from automated generation of timetables to driverless trains. For the purpose of this project, the authors have partnered with the Mass Transit Railway Corporation (MTR) Limited of Hong Kong.