Description:Field of Invention
The field of invention is combination of image processing and embedded systems. The present invention is appropriate to develop a reliable color sensing robot using image processing for the various real-time applications.

Objective of the Invention
Design and implement a reliable color sensing robot using image processing techniques for the various real-time applications such as defense and smart shopping .This system helps to reduce the human intervention in the border area and can controlled from remote places.

Background of the Invention
Color sensing is a most important process in image processing for various real time applications. MATLAB is one of the most important tools for perform operation in an image processing. It is a method to perform required operations on an image and extract the useful information. It include three main steps namely, image import, Examining and deploying the image and the third step is to produce an output image after performing manipulations on an image. Here, in this invention image is processed based on the color which is the most important characteristic of an image. Since image processing has led to the advancements in the field of robotics and embedded system this invention also includes design of electronic prototype which is at two wheeled robot.
(Vindhya D& et al., [2012], International Journal of Science and Advanced Technology, 9, p 30-35) has proposed “Design and Development of Object Recognition and Sorting Robots for Material Handling in Packaging and Logistic Industries”. In this proposition a MATLAB code is composed for the calculation for playing out the activity. The calculation is executed to distinguish the item and send the proper orders to the microcontroller involving sequential correspondence for the robot to play out the arranging activity. (Sanghoon Kim& et al., [2008], Second International Conference on Future Generation Communication and Networking, 1, p 78-85), A mobile robot with various types of sensors via ubiquitous networks is introduced. We designed a mobile robot composed of TCP/IP network, wireless camera and several sensors in an environment, and show object avoiding and tracking methods necessary for providing diverse services desired by the people.(Vijayalaxmi& et al., [2014], Global Journal of Advanced Engineering Technologies, 9, p -35),''object detection and tracking using image processing''. The rudimentary disclosure process exists of examining the portrayal lattice and at each positions try either Xs+W is top-secret as item or foundation. This is generally achieved at broadened judgments of the portrayal pyramid to find objects at differentiated scales, and is most certainly an exceptionally comprehensive estimation. There are any of techniques to oversee more old].

(Mr.Prakasha M& et al., [2017], International Journal of Engineering Science Invention Research & Development, 4, p 28-39), the proposed work incorporates distinguishing a proficient movement element to identify the moving articles. This model will be carried out utilizing MATLAB and execution of the framework is assessed.
Description of Prior Art
The robot that is configured to display a third image related to a second image when a first operation is accepted and a first image is picked up and the second image included in the first image is selected, to set a condition for a process on an object whose image is picked up when an information of the third image is accepted, and to carry out the process based on the set condition. ( US20180056516A1).The image processing incorporates a point of interaction that gets a majority of caught pictures got by the imaging unit, an estimation unit arranged to obtain an estimation after effect of a subject in the caught picture by performing estimation handling to the caught picture, an amalgamation unit designed to create an orchestrated picture by combining the majority of caught pictures to such an extent that the caught pictures cross-over with each other inside a covering range relating to the covering area in imaging request; and a result unit designed to yield the integrated picture, data showing the covering range connected with the blended picture, and data demonstrating the estimation result. (US20140015956A1). In a method for interactive marking by a mobile robot on a vertical surface, a mobile robot that includes a sensor and an actuated marker is displaced across a vertical surface. Features on, in or behind the vertical surface are detected with the sensor. Displacement of the mobile robot and actuation of the actuated marker is controlled in response to the detection of these features. (US20170036349A1). In a negligibly intrusive careful framework, a picture catch unit incorporates a crystal get together and sensor gathering. The crystal get together incorporates a bar splitter, while the sensor gathering incorporates coplanar picture catch sensors. Every one of the coplanar picture catch sensors has a typical front end optical construction, e.g., the optical design distal to the picture catch unit is no different for every one of the sensors. A regulator improves pictures gained by the coplanar picture catch sensors (US20130041221A1).
Summary of the Invention
The important feature of the robot is that physical movements can be controlled using MATLAB software. Controlling process is more accurate and cost effective because we use the MATLAB software. The robot is also cheaper than existing one and implemented using simple circuit when compared to the existing one. After the implementation of the invention is come across the results like forward movement, in reverse movement and left movement, right movement and stop. Barely any more highlights can be added to further develop the exhibition with the end goal that the robot can be prepared to stay away from the deterrents and various articles can be recognized in a solitary edge and furthermore numerous tones can be distinguished which is utilized in the applications like arranging in light of the variety. This invention can be extended by adding a robotic arm to pick and place the products of different color with flexible movement of arm. This can perform complex tasks. This invention is designed, implemented and successfully tested with real time test data. This invention finds applications in many fields like automation Industry, Industrial Applications, Automotive agro Applications, defense etc.

Detailed Description of the Invention
This invention uses a simple method of image processing which processes the image captured and senses the required color from that image and tracks the color which result in the movement of the robot based on color detected. The design of this invention consists of two different sections namely Image Processing section and Robot control circuit section. The application running on the computer detects the color of an object and interprets its position and sends the respective intermediate instructions to the micro controller through the serial port. The robot is located at the receiver end which receives those intermediate instructions from the computer and the motors are controlled through the motor driver IC.
Firstly the image is captured through the sensor which is a web camera embedded in the PC. Since the MATLAB software is already loaded in the PC the image processing technique is achieved through the MATLAB software which processes the image captured from the sensor.
Image processing can be used to enhance an image's quality, get rid of unwanted items from an image, or even make entirely new images. For instance, backdrop can be eliminated from a picture of a person using image processing, leaving the subject in the foreground. There are numerous different algorithms and approaches that can be applied to produce various results in the broad and complex field of image processing. The most popular image processing jobs and their execution will be the main topics of this section.
Image enhancement, or raising the quality of an image, is one of the most frequent jobs involved in image processing. It is essential for surveillance, remote sensing, and computer vision jobs. Changing the image's contrast and brightness is a typical strategy.
Images' quality may deteriorate for a number of reasons; especially if they date from a time before cloud storage was widely used. As an illustration, photos scanned from hard copies made with vintage instant cameras frequently develop scratches.
Based on the required color is detected or not the corresponding intermediate instructions are sent to the robotic circuit. Based on the instructions received the motors of the robot are driven by the motor driver IC and results in the movement of the robot. The power supply must be provided to the microcontroller and motor driver of 5v which is regulated using voltage regulator. ARDUINO NANO is a small, complete and breadboard friendly microcontroller which is based on ATMEGA328P.They are widely used in robotics, automation. This microcontroller has mini-USB port which is generally used for programming and serial monitoring. It has eight analog input output pins and 14 digital input output pins, 2 reset pins and 6 power pins. The input voltage supplied to this microcontroller can vary between 7 to 12 volts. The operating voltage is 5v.This microcontroller is programmed using ARDUINO IDE.

The L293D IC is also known as motor driver. It is 16 pin IC. It has four input pins, four output pins, 2 enable pins. It has two power supply pins to make a connection to the motors. It deals with the H-span idea. H-Bridge is a circuit which permits the voltage to be flown in one or the other heading. As need might arise to adjust its course for pivoting the engine of robot either in clockwise or against clockwise headings. H-bridge IC is ideal for driving a DC motor. An electrical circuit known as H-bridge changes the polarity of a voltage applied to a load. These circuits enable DC motors to move forward or backward in robotics and other applications. The main advantage of this IC is high noise immunity and over temperature protection is available.7805 IC is also known as voltage regulator. This regulator is a three terminal and linear voltage regulator. Since the power supply is 12v it is regulated to 5v to drive the microcontroller and motor driver IC. 2 pin bug connectors are used to provide the Vcc and ground supplies to the microcontroller and motor driver IC. It has positive and negative pins. LED issued to indicate the flow of current through the circuit. Resistor is connected to the LED to turn it ON. Capacitor is a noise in the process of instructions received from the PC. So to eliminate this noise capacitor of 1000uf value is used. Switch is generally used to switch on/off the robot circuit for further movements
Steps involved in color detection,
It extracts the frames from the input image.
After extracting the frames from image, it will be extracting the RGB color components.
Each frame is now converted into gray scale image.
Noiseiseliminatedfromtheimageusingamedianfilterwhichisafilteringtechnique.
The pixels which are less than 300px are eliminated otherwise which may result in malfunction.
Finally the resultant image is converted to binary image.
The algorithm ensures tracking of the object, If the object is found anywhere on the image plane.
This algorithm comprises of three algorithms namely
Color detection algorithm
This variety identification calculation distinguishes pixels in a picture that match a predefined variety or variety range. The shade of recognized pixels can then be changed to recognize them from the remainder of the picture.
Mean Absolute Error MAE= ?_(i=1)^M¦?_(j=1)^N¦?I(i,j)-g(i,j)?/M*N
Mean Square Error MSE= ?_(i=1)^M¦?_(j=1)^N¦?I(i,j)+f(x,y)?-b/M*N
Color tracking algorithm
Utilizing variety highlights in visual following is an appealing choice on account of its effortlessness and power under fractional impediment, profundity and scale changes. By the by, there exist a huge issues that should be addressed to plan a strong and exact variety tracker.
Motion control algorithm
A moving object/picture division technique in view of a better two-layered Otsu strategy is proposed to take care of the issue of incorrect and slow division of moving the article/pictures utilizing the two-layered Otsu strategy. Color detection and tracking algorithm is used to detect the object of the specified color and allows the user to make decisions where the object is to be moved based on our requirement.
Motion control algorithm shows how the robot must move. The robot will move in five
Directions, they are
Forward movement
Backward movement
Left movement
Right movement
Stop

Brief Description of Drawings
In the figure which is delineate praiseworthy exemplification of the innovation.
Figure 1: It shows the block diagram of color sensing robot using image processing techniques.
Figure2: Flow chart for Motion control algorithm
Figure 3: Flowchart for color sensing and tracking
Figure 4: Image of implemented hardware

5 Claims & 4 Figures , Claims:The scope of the invention is defined by the following claims:

Claims:
1. Automatic color sensing robot using image processing including the following system/method comprising the steps of:
a) Image processing is a front end of the invention. Simple method of image processing which perform various operation/ conversion on the captured image.
b) And Color detection algorithm senses the required color from that image and tracks the color which results in the movement of the robot based on color detected.
c) Embedded based Robotics is the backend of the invention. The robot is located at the receiver end which receives those intermediate instructions from the computer and the motors are controlled through the motor driver IC.
2. As mentioned in claim 1, frontend algorithms detect the color of the object in an image and backend algorithms helps to track the objects in an image.
3. According to claim 1, based on the detected color of the object in the image, robot can move four directions like forward, backward, left and Right.
4. As per claim 1, motion control algorithm controls the directions of the robot movements.
5. As per mentioned claim 1, the movements of the robot are controlled automaticity based on the colors detected by the receiver. Finally robot that can classify objects according to their color is proposed.