DOC PREVIEW
MIT 6 111 - Writing Recognition

This preview shows page 1-2-3-4-5 out of 15 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 15 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 15 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 15 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 15 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 15 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 15 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Writing Recognition Stephanie Hsu 6.111 Fall 2006 Final Project Abstract: Writing recognition programs are an increasing component of electronic tools such as personal data assistants and document processing suites. However, the challenges of recognizing a hand-drawn image stems from the myriad of ways each character can be depicted. The project aims to implement a form of writing recognition by constraining the user to a specific set of predetermined instructions and searching for the correct character based on the directional movement of the sketch. In this project, the user draws a character by moving a mouseContents I. Overview 1 II. Module Descriptions 8 III. Testing and Debugging 13 IV. Conclusion 14I. Overview The project aims to recognize a user-drawn letter or number. The user sketches the character by clicking and moving the mouse within the writing pad displayed on the monitor. The strategy to recognize the written character is based on techniques used by writing recognition software installed on personal data assistants and relies on a predetermined set of stroke rules for each character. 1.1 User Interface The user interfaces with the project through the mouse and the display monitor. A mouse cursor moved by the user is displayed on the monitor. The outlines of a writing pad and the recognized character are also shown on the display. Outlines of the areas on the display are shown in Figure 1. The characters the project can recognize are shown in Figure 2. The user must begin sketching the character on the dot and follow the stroke order displayed in Figure 2. Lower and upper-case letters are not distinguishable by the order of the mouse stroke. An upper case letter will be displayed if the right mouse-button is clicked while a lower-case letter will be displayed if the left mouse-button is clicked while the user is writing a character. In addition, the user must write the character within the writing pad outline on the monitor. Characters recognized displayed Figure 1: Outline of Display1.2 Character Recognition Algorithm The algorithm for recognizing characters relies on a specific set of instructions the user must follow to write a character. Each character begins and ends in a pre-determined area of the writing pad, which is divided into eight cells as shown in Figure 3. Based on the beginning and ending cell of a writing sequence, the algorithm assigns one of seventeen possible FSMs to attempt to identify the character. The assigned FSM analyzes the sequence of cells that the mouse travels through to match the sketched character with the correct ASCII code. If no character match can be found, the display will output a question mark. Figure 3: Writing Pad 0000 0001 0010 0011 0100 0101 0111 0110 Figure 2: Recognizable Letters and Numbers: The dot indicates the starting point of the mouseThis method serves to quickly divide possible matches into sections based on the broad criteria of the first and last cell entered by the mouse. Each FSM then sequentially analyzes the sequence of cells the mouse traveled through to identify the character written. A sample FSM that analyzes sequences beginning in cells 0 or 2 and ending in cells 1 or 3 is shown in Figure 5. 1.3 Implementation The project consists of 3 large modules, each of which is broken down into several smaller submodules. The connections between the modules are shown in Figure 6. The mouse module decodes data input from a PS/2 mouse interfaced with the labkit and translates the information into x and y position coordinates on the display screen. In addition, the module also outputs information about the buttons clicked on the mouse. The character recognition module stores information about the movement of the mouse when the user is writing by storing the sequence of cells that the mouse passes through into BRAM. The BRAM is accessed by the character recognition to implement the algorithm described in section 1.2. Once the character has been identified, the ASCII code is sent to the display module and output onto the monitor. A B C D E: Valid “V” F G: Valid “W” (2, 4) (4) (5, 6) (7) (3) (6) (5, 7) Figure 5: FSM 13, which decides between “V” and “W.” Numbers in ( ) are the input cells from memory that cause the FSM to change state. Numbers in [ ] indicate that the input cell is the last of the sequence. [1, 3] [1, 3] (5, 6) (3, 5, 7)Reset Mouse Module Mouse_data Left button x location Figure 7: Mouse Module y location Right button II. Modules 2.1 Mouse Module The mouse module receives data from a PS/2 mouse interfaced with the labkit, which transmits changes in the x and y position of the mouse as well as information about the mouse. The mouse module (modified example from the 6.111 code bank) uses the top right hand corner of the screen as a reset point and determines the actual location of the mouse on the screen. The x and y coordinates are output to the character recognition module. Figure 7 outlines the input and output signals of the mouse module. \ Mouse Module Left Button x location y location Right Button mouse input reset Character Recognition Module next_char find_finish Display Generator To monitor Figure 6: Block Diagram Major ModulesThe mouse module runs continuously and goes to the upper left-hand corner of the screen once the reset button is pushed. The module was taken from the Fall 2005 6.111 code bank [1]. 2.2 Character Recognition Module The bulk of the project, the character recognition module analyzes the position of the mouse on the screen, identifies the character the user is sketching, and outputs the ASCII code for the recognized character to the display module. Figure 8 illustrates the components of the module. BRAM: The 16x8 bit memory block stores the sequence of cells the mouse passes through while writing a character. mouse_cell controller: The mouse_cell controller uses the x and y coordinates extracted from the mouse input data to identify the area of the display the mouse cursor occupies. If the mouse cursor is not within the writing pad, mouse_cell is set to 1000. Otherwise, mouse_cell equals the number of the cell (as shown in Figure 3) it occupies. Mouse_cell is output to the mouse_finish controller and the BRAM. Parameters within the module determine the size of the writing pad. mouse_finish controller: The mouse_finish signal is only low if the


View Full Document

MIT 6 111 - Writing Recognition

Documents in this Course
Verilog

Verilog

21 pages

Video

Video

28 pages

Bass Hero

Bass Hero

17 pages

Deep 3D

Deep 3D

12 pages

SERPENT

SERPENT

8 pages

Vertex

Vertex

92 pages

Vertex

Vertex

4 pages

Snapshot

Snapshot

15 pages

Memories

Memories

42 pages

Deep3D

Deep3D

60 pages

Design

Design

2 pages

Frogger

Frogger

11 pages

SkiFree

SkiFree

81 pages

Vertex

Vertex

10 pages

EXPRESS

EXPRESS

2 pages

Labyrinth

Labyrinth

81 pages

Load more
Download Writing Recognition
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Writing Recognition and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Writing Recognition 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?