This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
5.1 Data Flow Diagrams
A Data Flow Diagram (DFD) is a graphical illustration of the “flow” of data through out an information system. The Data Flow Diagrams (DFD's) are used to show how data passes through a sequence of transforming steps. Before moving on to the next level the data is transformed at each step . A DFD depicts what kinds of data will be the output from and input to the system, where the data is going to and coming from, and where the data will be saved. It does not show about the timing of processes or about if processes will operate in parallel or in sequence. These transformations or processing steps are program functions, when Data Flow models are used to form a software design. The DFD diagram's are composed of four elements such as the process, data flow, external entity and data store. In the proposed method the DFD is divided into three levels they are as follows: level 0, level 1, and level 2.
Symbols of DFD
- External Entity
External entities are those entities which are identified as entities required to interact with the systems under study. The external entities are used to show output data from the system and input data to the system or both.
Figure 5.1 External Entity
Typically external entities may represent work titles or the system that interact with the other systems to be built. The external entity symbol is depicted as shown in the Figure 5.1.
The action that is carried out with the data that flows across the system is referred to as a process. A process uses input data required for the process to be taken place and produces data which is sent to next part of the DFD. Processes symbol is depicted as shown in the Figure 5.2.
Figure 5.2. Processes
- Data stores
Data store is referred to as the place where data or information may be stored. This information or data may be saved either permanently or temporarily by the end user. Files and tables are represented as data stores. A rectangle open on the right hand side is generally used to refer data stores. The data stores symbol is depicted as shown in the Figure 5.3.
Figure 5.3. Data stores
- Data Flow
Data flow is the flow of the data between the entities, the process and the data store. Data flow illustrates the connection between the components of the DFD.
Figure 5.4. Data flow
Data flow is represented by an arrow, where the arrow is commented with data name. The data flow symbol is depicted as shown in the Figure 5.4.
Data Flow Diagrams – Level 0
The initial level DFD is the level-0 and it is generally called as the context level diagram. A designer must draw a context-level DFD first in common practice, which illustrates the connection between outside entities and the system . The context-level DFD is break out to show more detailed design of the system. The Figure 5.5 shows the level-0 DFD of the face recognition system for differing pose and brightness, where the user gives face image as the input; Output obtained as whether match found or not found in the Database. The input image can be a grayscale or RGB image, if the image is RGB image then it is converted to the grayscale image and it is being processed.
Figure 5.5 the level-0 DFD
The system displays whether the input image is present or not in the database to the end user as an output.
Data Flow Diagram – level 1
The Level-1 DFD gives more detailed information than the level-0 DFD. The Level-1 DFD is as shown in firuge 5.6.
Figure 5.6 the level-1 DFD
The level-1 DFD consists of the following phases. They are image pre-processing, face detection and location of landmarks, pose and illumination normalization and face matching. The user gives face image as an input image. The input face image is given as an input to the image pre-processing step, the input face image is converted to grayscale image and removal of noise such as salt and pepper noise using the median filter is carried out in this step. The noiseless grayscale face image is given as an input to the face detection and location of landmarks step, where the ROI (Region Of Interest) of the face image is detected and position of feature points using STASM approach is carried out. The ROI of the face image is given as an input to the pose and illumination step, where the pose and illumination normalization is carried out. The normalized face image is given as an input to the face matching step, where face recognition is carried out using the matching algorithm.The Figure 5.6 shows the level-1 DFD of the face recognition system which consists of the image processing, face detection, pose and illumination normalization and face matching.
Data Flow Diagram – level 2
The Level-2 Data Flow Diagram gives more information than the level-1 DFD. The following figures show the Level-2 DFD for each module.
The Image pre-processing consists of the conversion of a RGB image into a grayscale image, resizing of an image and removal of noise using median filter. The level-2 DFD for image pre-processing is shown in the figure 5.7.
Figure 5.7 the level-2 DFD for image pre-processing
FACE DETECTION AND NORMALIZATION:
The face detection and normalization consists of detection of ROI in the input face image, location of landmarks and pose and illumination normalization. The Figure 5.8 shows the complete steps involved in Face detection and normalization processes are as mentioned in the diagram below:
Figure 5.8 Face detection and normalization process
Figure 5.9 Recognizing the face image
The Figure 5.9 explains the block which performs the function of recognizing the input image with the image present in the Database. It performs 1:N matching using Matching Algorithm.
5.2 Use Case Diagram
The use case model acquires the conditions of the system. Use cases are the means for communicating with other actors and end users regarding what the system is intended to do. Use Case Diagrams depicts a set of use cases, actors and their relationships. A use case diagram depicts the interaction between the system and external entities to the system. The external entities are referred to as actor's. The use case diagrams depicts the static use case view of a system and are important in organizing and modelling the behaviours of a system. The use case diagram for face recognition system for differing pose and brightness is depicted in the figure 5.10.
Figure 5.10 Use Case Diagram
The use cases for the face recognition system are as follows - Read face image, Image pre-processing, Face detection, Location of landmarks, Normalization, Face matching. The user reads a face image with differing pose and brightness from the system database and he gets the result as whether the selected face image is present or not in the system database.
5.3 Sequence Diagram
A sequence diagram is referred to as an interaction diagram which depicts how processes interact with each other and the order of the processes. An object interactions arranged in time sequence is depicted in sequence diagrams. Sequence diagrams shows the object and classes involved in the sequence of steps and the array of messages interchanged with the objects. Sequence diagrams are also referred to as event diagrams.
Figure 5.11 sequence diagram for face recognition system
A sequence diagram is represented as parallel vertical lines illustrating the process or objects and horizontal arrows illustrating the messages exchanged between the process or objects in the order. Sequence diagram for face recognition system for differing pose and brightness is as shown in the figure 5.11. The different processes or objects in the face recognition system are user, image pre-processing, location of points, normalization and face matching. The user sends the face image as the input. The Face matching process searches for the input face image (1:N matching) in the system database and sends the user match found or no-match found to the user.