Research on Detecting Hand Gesture

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 09 Issue: 04 | Apr 2022 www.irjet.net p-ISSN: 2395-0072
© 2022, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 1370
Research on Detecting Hand Gesture
Abhishek Nair M.B1
, Priya N3
1
PG Student, Department of Computer Science, Jain University, Bangalore, Karnataka. India
3
Professor, Department of Computer Science, Jain University, Bangalore, Karnataka.India
---------------------------------------------------------------------***--------------------------------------------------------------------
Abstract - This project presents a way to improve real-time visual perception using only webcams and computer vision
technologies such as multi-contact detectable image processing for use in computer interactions. almost everywhere we
interact with computers. An important feature of this project is to simulate mouse as a visual input device with all its
functions like left click, right click
1. INTRODUCTION
Today, human computer interactions are mostly interactions with nature. The Graphic User Interface (GUI) on the
Personal Computer (PC) is being upgraded, providing an effective visual interface for users to interact with the computer
and easily access various applications. Today the technology used for mobile communication is especially the touch
screen. But these technologies are still not cheap enough to use on laptops and desktops. We are building a virtual mouse
system with easy to use interface and different touch screen. Visual devices (web cameras) may be used and upgraded to
new input devices in the future. Provides the installation command for the computer with a computer connection. We can
implement additional implementations to convert computer vision tools into an input control device to access a function
such as a keyboard or mouse. One way to signal visual aids on a computer is through touch. Live hand gestures are used
as signals or input methods. A particular signal can be detected by the computer as input by the computer. This will
benefit every user who doesn't need to use a simple device and can do whatever they want as long as the computer vision
device can hear you. This makes using the computer easier than using a keyboard or mouse. A future computer or laptop
could eliminate the use of a keyboard and mouse in exchange for vision-based translation tools. Human interactions
occur in different sensory systems such as touch, speech, facial expressions, and body language. The great advantage of
using hand gestures is interacting with the computer as a personal computer input method. Our application provides
efficient and easy-to-use methods for intelligent computer interaction and using hand gestures. Mouse functions such as
intuitive object movement control have been replaced by hand gestures. Complexity exists consistent with the acquisition
and recognition phases of a simulated application. The challenges faced by noisy environments produce severe
impairments in the acquisition and functioning of touch in the human hand. The app is designed to be inexpensive and
uses inexpensive input tools like webcams for hands as input. Material manipulation is done by means of a preconceived
notion based on hand gestures.
2. AIM OF THE PROJECT
The main goal is to create an intuitive mouse using a touch screen camera so that we can interact with the computer and
easily access different applications.
3. SCOPE
 For most laptops, the touchpad is uncomfortable and not convenient
 Real time application
 Easy to use
4. PROBLEM STATEMENT
 Visual-based mouse design receives wearable models instead of visual mice.
 Basically for this project using hand gestures recorded by a webcam.
 The camera captures and detects hand movements and performs mouse functions.

5. RELATED WORKS
5.1 Virtual mouse using gesture hand
 Suggested cursor control system for camera view, using hand gestures taken from webcam by color detection
method.
 The system will allow the user to navigate on the computer screen using a hand-held hat or color strip and left-
clicking and dragging will be performed with another hand gesture.
 Hand recognition uses a simple color on the finger to operate the pointer using basic hand and touch commands
without additional hardware. This is done through the use of intuitive touch recognition and webcaminput.
Advantages - real time application
Disadvantages- User needs color tapes or caps
5.2 Virtual Keyboard
The virtual keyboard, or "on-screen" keyboard, allows you to easily and seamlessly type your local language scripts
directly from anywhere on whatever computer you're using.
Other common uses of the virtual keyboard include:
 Allow a person to type in their own language on external keyboards
 Provide users with more accessible input by allowing on-screen typing,
 Provides a quick and easy way to switch between different character sets and/or alphabets.
5.3 Touch Music Player
The touch player can be used to play songs with hand gestures. That means no keyboard or mouse needed. Touch can
include play, pause, next, past, volume up, volume down, etc.
5,4 Visual Pian
Intuitive piano functions similar to a virtual keyboard. One can play the piano directly with the help of hand gesture from
the card to the corresponding key holding the corresponding piece of music.
5.5 Visual Artist
With just one touch, we can draw on the screen. That helps us facilitate our efforts. We can draw the screen easily by
touching the body. Traditionally, we use an intuitive mouse. For example, in the window with paint, we use the visual
mouse to draw the image. By using visual artist we can easily draw objects with hand movements.
6 ANALYSIS AND INTERPRETATION
Input devices such as old mice and peripherals offer only a limited number of them, and they are mainly used for gaming
purposes. Legacy devices are finished products or hardware. This is the main reason for the high cost of the product.
Product maintenance is also very expensive. The goal of this project is to implement hand tracking not only for gaming
purposes but also for routine use. Since we only use one web camera, costs can be significantly reduced.
6.1 EXISTING SYSTEMS
6.1.1 Microsoft Kinect
Kinect is a controllerless gaming experience for the Xbox 360 video game platform. It relies on a webcam-style peripheral
addon to interact with the Xbox 360. It allows users to browse and interact. with Xbox 360 without touching a game
controller through a user interface that uses gestures, voice commands, or certain objects and images.

6.1.2 Leap motion controller
The leap Motion Controller is a small USB device that plugs into a computer. Using an LED light and camera sensor, the
leap Motion Controller scans an area approximately 8 cubic feet (2 x 2 x 2) above the device. It tracks both your hands
and 10 fingers as they move through the open space between you and the computer, special software detects your hands
and fingers and converts the data into information for the computer. yours.
Drawbacks
In Microsoft Kinect, the big drawback is that it requires too much space. Cost is another matter. The most frustrating part
of the kinect is the menu system at your fingertips. The built-in Xbox menu interface lets you hover over menu blocks for
a few seconds before anything happens.
The problem with leap motion is that it follows our 10 fingers if one finger overlaps the other then there is a motion
tracking problem. Actual performance may not translate to computer model as expected
6.2 PROPOSED SYSTEM
The proposed program is vision-based, using techniques to process and capture images from a web or digital camera. The
input frame will be captured on a digital camera, and the system is divided into four stages, skin detection, manual
contour removal, skin area to be detected, skin trauma detection. Then the outline of the hand will be detected and used
for hand tracking and touch detection. Manual tracking will be used to navigate to PC ID and manual touch will be used to
perform mouse functions such as right click, left click, scroll up and scroll down. Therefore, the scope of the project will
be the design of a vision-based system.
6.3 METHODOLOGY
The main contribution of the recommendation system consists of four parts:
Skin Detection - 5 skin color sampler using grayscale values
Contour Extraction - Extend the concert of hands using histogram values
Hand tracking and touch recognition - Hand tracking using clear image points and algorithms
Cursor controls - Mapping touch control to pointer control
Acquiring skin using color detail can be a daunting task, as the appearance of skin in photos is affected by lighting,
camera features, background, and quality. To reduce the influence of light, images can be converted to color palettes,
which are less sensitive to changes in light. After finding a binary image delimited by skin elements, edge detection is
performed to find the gigs in the image using the OpenCV cvFindContours() function.
Cursor movement is controlled by the tip of the finger. To see the fingertips, you must first find the center of the palm.
The shortest distance between each point in the circle is engraved into the measured lance and the point of the largest
distance has been recorded as an organization. The distance between the navel of the hand and palms is considered the
width of the hand. The center of the hand is calculated throughout the sequence and using the manual center, when
discovering the hand gestures, it will be a simple question about hand-edged gestures completely different from functions
Specific mice. It seems to control your computer pointer.
RGB image - RGB images, sometimes called TrueColor images, are a series of data that puts red, green and blue
components for each individual pixel. A pixel with color components of (0,0,0) displays as black, and a pixel with color
components of (1,1,1) displays as white.
HSV image consists of three parts
“Hue” is more specifically described by the dominant wavelength and is the first factor we refer to when adding the three
components of a color. Hue is also a term that describes the one dimension of color that we easily experience when
looking at color, or its purest form; it basically refers to a color with full saturation

Value - The value refers to the lightness or darkness of a color. It indicates the amount of reflected light. When dealing
with pigments, dark values that add black are called shades of the given color name. Light values with the addition of
white pigment are called hues from the color name. The value just refers to the lightness or darkness of a color. If a color
is called "low value" it means it is darker in nature think closer to earth (dark ground) and if it is called "high value" it
means it has obviously darker nature think closest to the whitest, brightest sky.
Saturation - Saturation determines the brightness and intensity of the color. When a color is tinted, white and black
(gray) are added to the color to reduce the saturation of the color. When a color is highly saturated, the intensity of the
pigment is vivid and bright. When a color is desaturated or unsaturated, the color appears "grayed out" or muted.
Grayscale images include two important pieces of information: the change in grayscale and the local geometrical
features of the image. A primitive sketch is a primitive mode of expression that can fully and unambiguously represent
information. Much of the information in the early sketches focuses on the rapid change in grayscale from the actual edges
and ends. Each edge grayscale change has a corresponding description, including edge grayscale change rate, total
grayscale change, length, curvature, and edge orientation. Erosion and dilation
Erosion and dilation are morphological image processing operations. OpenCV morphological image processing is a
procedure for modifying the geometric structure of an image. In metamorphosis we find the shape and size or structure
of an object. Both operations are defined for binary images, but we can also use them on grayscale images. They are
widely used in the following ways:
 Noise Reduction
 Identifies any intensity bumps or holes in the image.
 Split individual elements and place different elements in the image.
Smoothing and thresholding are used to see the useful part of the image, while smoothing is used to remove noise from
the image. The images captured by the camera are usually in an analog format and are therefore digitized by sampling.
The scanned images are stored in the computer as arrays of pixels. Each pixel contains information such as the color or
intensity of that part of the image. The pixel size depends on the sample size of the scan. In image thresholding, each pixel
value is set to black or white based on a user-defined threshold value. When smoothing, the pixel value is set based on the
value of the neighboring pixels.
6.4 BLOCK DIAGRAM
7 CONCLUSION
In this project, a track-based mouse application was developed and deployed using a webcam. The program is executed
on behalf of PYTHON. We implement mouse functions like left and right click. Most applications require additional, often
more expensive, hardware. The aim is to make the technology cheaper and older under the standard operating system.
Various application programs can only be written for this technology to create many types of applications with minimal
resources.

REFERENCES
[1]Chen, FS, Fu, CM and Huang, CL. (2003). Recognition of hand gestures using real-time tracking and hidden Markov
models. Imaging and Computer Vision , 21 (8), 745758.
[2]Dipietro, L., Sabatini, A. M., & Dario, P. (2008). A survey of glove-based systems and their applications. That is,
transactions on systems, people and cybernetics, part c (applications and evaluation), 38 (4), 461 482.
[3]Dong, G., Yan, Y., & Xie , M. (1998). Vision-based hand gesture recognition for human-vehicle interaction. Contact
information presented with Proc. of the international conference on control, automation and computer vision.
[4]Garg, P., Aggarwal, N., & Sofat, S. (2009). Vision-based hand gesture recognition. World Academy of Science, Engineering
and Technology, 49 (1), 972977.
[5] Gupta, S., Jaafar, J. and Ahmad, W. F. W. (2012). Static recognition of hand gestures using a local Gabor filter. Procedural
Engineering, 41, 827832.
[6] Hasan, H., & AbdulKareem, S. (2014). Retracted article: Human–computer interaction using visionbased hand gesture
recognition systems: A survey. Neural Computing and Applications, 25(2), 251261.
[7] Hasan, M. M., & Mishra, P. K. (2012). Modeling and recognizing hand gestures using geometric features: evaluation.
Canadian Journal of Image Processing and Computer Vision , 3 (1), 1226.

Research on Detecting Hand Gesture

More Related Content

Similar to Research on Detecting Hand Gesture (20)

More from IRJET Journal (20)

Recently uploaded (20)

Research on Detecting Hand Gesture