Beta's projects for final year students 2012–2013

General

Dealiasing

1 student, maximum 2 groups | Research+Development project | Under the "Cooperative training programmes" between HKU CS and HKO

Description

A Doppler radar give at least two sets of data in polar coordinates: the reflectivity, and the Doppler velocity. The former is related to the strength of the reflected signal, and thus the amount of precipitations (e.g., rain, snow) in the space being scanned. The latter is related to the radial speed of movement of the precipitation.

A brief introduction to the operation of radars can be found at the HKO page on "What is a Weather Radar?". Sample images of the radial velocity field can be found on the page on "1994" in the 50th Anniversary of Weather Radar Observations in Hong Kong Commemorative Booklet. The images shown on the "Weather Radar Images" tab group at HKO's web site Weather Radar Image are reflectivity data at 3km Constant Altitude Plan Position Indicator (CAPPI).

Since Doppler radar works by sensing the reflected pulses it periodically sends to the air, there is limitations on the radial speed of the precipitations it can measure. When the radial speed of the object exceeds the limit, it will report a radial speed wrapped around to its lowest range. For example, if the range of a Doppler radar is from -10 m/s to +10 ms, and the precipitation is moving at +12 m/s, it will report the speed as -8 m/s, as the "extra" 2 m/s is wrapped around and (-10 + 2) m/s = 8 m/s is reported. This is called aliasing. Multiple folds of aliasing is possible. (what will the radar above report if the speed is 23m/s?)

Methods of Dealiasing are abound. Student taking this project is expected to:

Better students will be able to handle dual-radar problem which takes data from two radars to improve the dealiasing performance and put it into real life use.

Requirements

Deliverables

iOS-based image/video analysis app

2 to 3 students, maximum 2 groups | Development project

Description

The idea is to develop iOS apps that analyzes what's taken from the camera of an iOS device to create responses. Some ideas include:

The student should give the initial idea of the application, which is expected to be refined in regular meetings, similar to the storyboard discussions of the multimedia course.

When you consider what to develop, remember to keep the simplicity of mobile apps; it's okay to develop multiple apps with the same theme in a project (the theme is important), or make them officially separate projects, rather than developing a swiss-knife type of app. It's always good to develop apps you consider interesting or needed, rather than being driven by the topic of the project.

Requirements

Deliverables

iOS and Android united

3 to 4 students, maximum 2 groups | Development project

Description

There are a number of tool kits and libraries that allow cross-platform app development for iOS and Android devices. A number of them are JavaScript- and/or HTML5-based. Few generates platform-specific source codes for their respective development platforms (specifically, XCode and Eclipse).

The project is about developing a set of tools that runs on a POSIX platform for cross-platform app development. The platform to run the tools is preferably OS X, but Linux is also possible. Windows aficionados should document the way the environment is set up in detail. The tools developed should generate platform-specific source, and yet allowing platform-specific code to be added. The method should be more elaborate than a set of tools for #ifdef kind of macro processing. The output of the tool kit or environment should be sets of source code and resources for direct compilation under XCode (for iOS) and Eclipse (for Android). The tools should allow platform-specific customization so that no or minimal change of code after code generation is needed. Design of a metalanguage may be needed so code for a task (e.g., use of location sensors) can be generated in both platforms without need to write things multiple times.

The student is expected to do some research on cross-platform app development tool kits or environments, and have some insight on how their development can be united. It is possible for students to opt for supporting only APIs of selected versions (e.g., iOS5, Android SDK version 16) or newer.

Requirements

Deliverables