Previous Home Next

Requirements and Specifications

3.1              Fictitious Scenario

For the purpose of simplifying the design analysis of the proposed project, a fictitious scenario was assumed as follows:

 

An enthusiastic architectural conservationist possesses a valuable archive in the form of measured drawings, photographs, textual data of many heritage structures, which his firm has gathered and prepared over a period of 25 years. More recent measured drawings are in the form of CAD drawings, digital photographs, and electronic documents (consisting the history of the structure and the reconstructions carried out by his firm).

 

The conservationist needs to refer to a lot of details if any restoration or reconstruction has to be carried out on a particular structure. He follows the only method that is available to him for finding the references between the measured drawings, photographs and textual information, and that is by usage of natural language (using alphanumeric characters). He finds it very cumbersome and time consuming to check and cross-refer every detail for planning out his professional itenary. Looking at the major developments in the field of computing and virtual reality, the conservationist decides to investigate the possibilities with a software professional in the graphics field.

 

A discussion with the software professional follows. A list of items required and future possibilities of the project are discussed. The conservationist and the software professional come up with the following ideas for the proposed application:

-          The primary aim of the application would be to link together all related information of different formats pertaining to a particular heritage structure and to make an application which facilitates referring to this information.

-          The application would be built in 3 stages. First would be to make a prototype of a referencing tool with very simple functions. Second would be to add advanced functionality to the referencing tool. And finally the third stage would be to create a “creation and editing tool” for creating new archives or modifying existing ones.

-          The application would have to make extensive use of 3D technologies as a basis for creating references to other data formats.

-          The application would be user friendly to an extent that even a non-professional could use it for research purposes. This is necessary as Historians or even conservationists may not understand any software complexities. Furthermore, this tool can then be made available to general population or to the tourist industry for commercial exploitation.

-          The application would require dual or even triple experts (a person knowing the fields of architecture, 3D-modeling and software) for creating or editing the new file format.

The conservationist gives the “go ahead” for the first stage of the prototype and asks the software professional to chalk out a business and time plan.

3.2       Mission restated (Top)

To provide a referencing tool for “architectural heritage conservationists” and “the general population”, which facilitates easy referencing between existing diverse formats of documents, using new media and virtual reality.

 

The design of the system was split into three parts for convenience:

-          Features to be implemented in three months for the MSc. individual project. This was considered to be the first stage of proposed project (See figure 3.2 – top half).

-          Possible advanced features that were needed for a fully functional referencing application that can be used by the conservationists (See figure 3.2 – bottom half).

-          Features of a further advanced feature, the editing tool, for editing and creating the file format of the data of heritage structures suitable for the referencing application (See figure 3.3).

Figure 3.1: Schematic Diagram of User Interface for the Referencing Tool (Location and Ortho viewports are advanced features that are proposed to be implemented in second stage)

 

Note: The features are presented using standard Unified Modeling Language notations, the use case diagrams. For each use case, a detailed description is provided to give the specifications of the proposed system. A schematic diagram of the user interface for the referencing tool is shown in figure 3.1 for convenience of explanation.

3.3       Requirements Capture (Top)

Figure 3.2: Use Case Diagram showing features that were to be implemented and possible advanced functions.

 

Figure 3.3: Use Case Diagram showing possible functions that can be incorporated in an editing tool

3.4       Specifications (Top)

Use Case Description for the Referencing Tool (Refer to figure 3.2):

 

Features that were to be implemented for first stage:

Navigate in the virtual world

The user can explore the heritage structure in real time by navigating in this viewport using the keyboard arrow keys or the mouse.

Find distance between two selected points

The user can select two points (vertices) in the navigation viewport and request the distance to be generated in the metric system (mm). The following values will be generated – actual distance, angle in XY Plane, angle from XY Plane, distance on the X-axis, distance on the Y-axis, and distance on the Z-axis.

Examine selected part of model

The user can select various parts of the model in the navigation viewport and open it in the examination viewport to examine (by rotating) that part of the model. This part of the entire model will have the highest level of detail (LOD). At a time, only one part can be opened in the examination viewport.

View material information on selected part of model

The user can select various texture areas in the model to view textual details on the materials used for construction of the structure in reality.

View images

The user can sequentially view through the complete database of available images (photo-documentation) of the structure in the image viewport.

Search for textual information by category

The user can search for any textual data about the structure by specifying a category (e.g. Cultural history, material details, restoration and reconstruction history, etc).

Resize viewports

The user can resize (only maximize) the image and text viewports (ortho viewport in advanced features) for concentrated studies (e.g. focus only on textual information or focus only on image based data).

 

Possible advanced features (second stage):

Jump to any point in location viewport

The user can jump to any location by clicking a desired position in the location viewport. The current position of the user will be displayed in real-time over a wireframe top view of the structure.

Generate wireframe ortho projections

The user can request orthographic projections (plans and elevations) in the ortho viewport, zoom in, zoom out etc. The orthographic projections are stored in the database and not generated in real-time.

Generate wireframe cross section

The user can specify a cutting plane either in the navigation viewport or the ortho viewport to generate a wireframe section. The section will be generated in the ortho viewport. The user will be able to save this section in the database as another orthographic projection.

Find Geodetic data at selected position

The user can select any point in the navigation viewport and find geodetic information (mean sea level, latitude, longitude etc.)

Find other measurements

The user can select points in the ortho viewport or the navigation viewport to find distances and areas, radii, center points of curves etc.

View audio-visual data

The user can sequentially view through the complete database of available audio-visual data of the structure in the image viewport.

 

Table 3.1 Use Case Description for figure 3.2

Use Case Description for the Editing Tool (Refer to fig 3.3):

 

Possible advanced features (third stage):

Create accurate models

The editor can create accurate models by using an advance modeling tool or importing support for VRML format into the editing tool. A plug-in for industry standard modeling tools like AutoCAD, 3D Studio MAX, ArchiCAD etc. can be used to directly support exporting the 3D Data (lights, textures, cameras and mesh) into the file format supported by the editing tool. An accurate 3D model can be also produced by other techniques like photogrammetry and 3D laser scanning.

Create associations and references

The editor can create associations and references between the relevant 3D objects and the data of different formats (i.e. textual information, audio, video and images).

Create a database

The editor can store all the data (i.e. 3D model, textual information, audio, video and images) into the provided database.

Create an integrated file

The editor can use the database to form the archive format readable by the referencing tool.

Edit existing archive

The editor can load an existing archive and edit all its components (3D model, references and other data) and save the changes for use in the referencing tool.

 

Table 3.2 Use Case Description for figure 3.3

3.5       Selection of structure (Top)

The following conditions were considered while deciding the heritage structure for the project:

  1. Exterior Architectural complexity in terms of intricate details: For the purpose of the prototype project (first stage) focus was on creating a good referencing tool for current Heritage data. Hence a structure with average complexity was to be selected (also see item 3).
  2. Availability of accurate measured drawings: The primary basis of the referencing tool was orthographic projections (architectural measured drawings) from which accurate model could be created. Hence availability of these drawings was of vital importance.
  3. Estimated Time required to construct a 3D model: As the focus was on creating a referencing tool and not a 3D engine for rendering, the selected structure did not have too many intricate details. Maximum time was to be devoted to implementation of the referencing tool.
  4. Availability of textual description, written histories and photographs: Besides measured drawings, textual description, written histories and photographs were also required to show the working of the prototype.
  5. Time required to create a database: All the information related to the selected heritage structure was to be stored in a database for organized access. For the purpose of the prototype project, if the details comprise of too many details then, only those details, which effectively explain the working of the prototype, were to be used.
  6. Proximity: The structure had to be accessible and within easy reach for immediate site visits, if required.

 

Four options were considered:

  1. Sheffield Town Hall,
  2. Sheffield Cathedral,
  3. St. George’s Church – Flats, Sheffield,
  4. Wells Cathedral, Wells.

 

Considering the conditions, “Sheffield Town Hall” was considered as too complex and would consume majority of the time for the construction of the model. Sheffield Cathedral was rejected for the same reason. Wells Cathedral, besides being complex, is located too very far away. Hence “The St. George’s Church – Flats” seemed to be the best option. As the structure belongs to the University of Sheffield, orthographic projections were easily available from the University’s Estate Department (through Mr. James Ingram). Arrangement was yet to be made for acquiring photographs, and textual data. The structure is located within a few minutes walking distance. Hence, immediate site visits could be made easily. The structure is of heritage importance and is of moderate complexity and was considered appropriate for the first stage of the proposed project.

3.6              Design Analysis (Top)

The business class diagram gave an initial overall picture of the proposed application. These were changed during the implementation design depending upon the hardware and software capacities and optimization issues.

 

For clarifying the functions of each method in the classes, a description is also provided.

 

Figure 3.4: Business class diagram

 

Description for methods

openFile() will load the 3d-Model (using loadModel() of NModel class), the text (using loadText() of NText class) and the images (using loadText() of Image class) in the memory. Simultaneously, it will display (a) the model in the navigation viewport with the camera position defaulted to the first camera position (using displayModel() in NModel class), (b) the introductory text about the Heritage structure in the Text viewport (using displayIntroText() in NText class) and (c) the first image in the Images viewport (using displayFirstImage() of Image class).

 

findDist() will allow the user to select two points in the Navigation viewport by using selectTwoPoints(), calculate the distance using calcDist()  (both of NModel class) and display the resulting distance in the Text viewport using dispDist() (of NText class).

 

viewDetail() will allow the user to select allowable parts (chunks) of the 3d-model in the navigation viewport by using selectDetailToView() (of NModel class) and display the detailed model in the Examination viewport using dispDetail() (of Examiner class).

 

ViewTextureTextDetail() will allow the user to select faces (one at a time) of the 3d-model in the Examintation viewport by using selectTexture() (of Examiner class) and display the textual description of the selected texture in the Text viewport using dispTextureTextDetail() (of NText class).

 

viewNextImage() will allow the user to view all the available images in a user controlled slide show fashion. displayNextImage() (of Images class) will be responsible for displaying the images in the Image Viewport.

 

searchTextInfo() will allow the user to initiate a simple searching engine. displaySearchTypes() (of NText class) will display the options and displaySearchResult() will display the search result in the Text viewport.

 

toggleTextDisplay() will allow the user to toggle between whether to show or not to show relevant textual information and while navigating. If it is set to “Yes”, relevant textual information will be displayed in the Text viewport using the displayRelevantInfo() (of NText class) depending on the location of the user in the Navigation viewport. The location is determined by detectUserLocation() (of NModel class). Default will be set to “yes”. (A similar feature can be added for display of relevant images in the Images viewport but this is left for future enhancement)

 

toggleViewportsActivity() will toggle between the Examination and Navigation viewport. Due to hardware limitation, it is not feasible for two real time viewports to function smoothly. Hence this method will make sure that only one of the two viewports are active at a time.

 

Note: XML will be used to organize and store the different formats of related data. Rules and standards (for organization of this data and the methods of building the 3D-models) will be developed during the process of implementation. The final organization of data will become the prototype file format, which can be used with this referencing tool.

3.7              Hardware Requirements (Top)

The following hardware would be required for development and testing of the proposed project:

-          Pentium-III class microprocessor

-          128MB RAM

-          16MB VRAM

-          20 GB Hard disk

-          SVGA supporting High Colour (16 bit) at 1024 x 768 screen resolution

-          Other standard input devices (mouse and keyboard)

-          CD-ROM 36X or above

 

The following hardware would be required for running the first stage of the proposed project after implementation (henceforth referred to as “Target machine” or “Target system”):

-          Pentium-III class microprocessor

-          128MB RAM

-          16MB VRAM

-          15.3GB Hard disk with 300MB free space

-          SVGA supporting High Colour (16 bit) at 1024 x 768 screen resolution

-          Other standard input devices (mouse and keyboard)

-          CD-ROM 36X or above

-          Multimedia Kit (only for further stages).

3.8              Software Requirements (Top)

The following software would be required for development and testing of the proposed project:

-          Windows NT 4.0 or 98 or 2000pro.

-          JAVA Development kit (JDK 1.3.1) with VRML and XML support

-          Java3D API (SDK 1.2.1)

-          Virtual Reality Modeling/Markup Language (VRML)

-          Extended Markup Language (XML)

-          AutoCAD and 3D Studio MAX

-          Adobe Photoshop or any other Image editing software

-          Microsoft Word

 

The following software would be required for running the first stage of the proposed project after implementation on target machine:

-          Windows 98 or 2000pro.

-          JAVA Virtual Machine (JVM – JRE 1.3.1) with Java3D support.

-          Installation files of the first stage of the proposed application.

3.9              Testing and Evaluation (Top)

Testing would be carried out in two categories. First would be specification based testing (black box), whereby it would be checked whether the system caters to all the requirements mentioned in the use cases (figure 3.2 top half). And second would be usability testing, whereby the target users (one conservationist/ architect if possible, one student of architecture and a person without related background) were to be introduced to the system and their responses were to be noted.

 

Seven use-cases were proposed for implementation. Out of these, five features (i.e. user navigation, dimension queries, examination of selected part of model, view material information on selected part of model and search for textual information by category) were considered vital to demonstrate the importance of this system. Additionally the user interface was also a key issue. Hence, successful implementation of these features and good responses from the usability test group was decided as the criteria to determine the success of the project. A fair amount of effort was to be put in to implement all seven use-cases.

 

The selected heritage structure has a big database of information in terms of text, modeling and images for accurate representation. Since, the aim of the project was not to demonstrate accuracy of data, partial data from each type that is necessary for demonstrating the above-mentioned features, was decided to be used (as detailed in section 3.5).

 

Previous Home Next