Field
This invention is based on the fields Image Processing, Internet of Things and
Android Application used forfinding the vacancy slot and provide user convenience and
time consumption.
Background of the Invention
In recent years, the increase in the number of private and commercial vehicles have led to a progressive worsening related to traffic problems and parking availability, especially in highly populated urban areas. An experience of finding a vacant parking slot can be very stressful in densely populated areas, especially in peak hours and contributes largely to total traffic congestion and increases gas emissions in urban and overpopulated cities. With the emerging problem of parking cars, the ordinary parking system that does not provide any information about available parking areas would not be able to handle the problem effectively. Such parking process takes a long time, wastes significant gasoline, and emits extra vehicle exhaust that harms the environment.In the before proposed ideas people use the IR sensor, ultrasonic sensor and Microwave radar sensor to detect the vacant slot.
IR sensor consists of two parts light emitting diode and receiver: When a car is parked close to the sensor, the infrared light from the LED reflects off the car and is detected by the receiver. The received signal is converted into current and the information is passed to the arduino. Based on the vacant slot the LED is blink on outside the parking slot.Ultrasonic sensor emits sound waves to detect the car. If the car is found the information is passed to the relevant device. As like the IR sensor the further process is proceed as the need of the user.

These sensors have both advantages and disadvantages. The IR sensor consumes lots of power and it is very sensitive to environment changes. As like the IR sensor ultrasonic sensor also consumes lots of power and very sensitive to environmental changes. Microwave radar which do not easily affected by environmental condition. It can able to detect the moving and stationary object. But it is quite expensive in parking area there are many slot so, it is not affordable to fix in all the slots.
Current parking slots are based on human monitoring or sensor based system, the problem faced areless efficient due to manual monitoring, sensor are sensitive to surrounding condition and time consuming. To overcome this problem we are using the idea of mobile based application. The simple smartphone is used as a gateway and connectivity enabler. The vision based system developed the alternatives to the ordinary system to detect available parking location. In this project we are using RFID reader we are detecting the RFID tag in the car, based on the detecting we can deduce the money digitally without the use of man power. Using surveillance camera free parking area are detected and updated in application. The camera captures the video of the area which is then processed for detecting the vacancy. The mobile application provides the details of number of vacancy slots and number of slots filled in the parking area. The LCD display fixed at the entrance provide the total amount.
Objects of the invention
This project to overcome a situation of monitoring and managing a parking area using a vision based automated parking system. With the rapid increase of cars the need to find available parking space in the most efficient manner, to avoid traffic congestion in a parking area, is becoming a necessity in car park management. In this project, a new method and a mobile application running on deep learning and PC-based are presented to solve the problem of searching for parking spaces in the metro city. With the mobile application, the user can dynamically access the model developed using the deep learning method and the real-time and forward-looking results can rapidly be shown to the user.

The mobile application provides the details of the vacancy slot availability at the preferredlocation which helps the driver to know the parking slot by reducing time and eliminates traffic congestion occurred due to searching.
Summary of the Invention
The proposed parking system suggests an IoT based framework that sends information around free and involved parking places by means of mobile application. The perfect of making a smart City is presently getting to be conceivable with the rise of the Internet of Things. One of the key issues that cities relate to is car parking facilities and traffic management systems. In present day finding an accessible parking spot is continuously troublesome for drivers, and it tends to ended up with ever expanding number of private car users. This circumstance can be seen as an opportunity for smart cities to undertake actions in order to enhance the efficiency of parking resources thus leading to reduction in searching times, traffic congestion and road accidents. Nowadays there is availability of parking details in real time. Such frameworks require efficient sensors to be sent within the"parking zones for observing the inhabitancies as well as quick data processing units in order to gain practical insights from data collected over . various sources. Issues relating to parking and traffic congestion can be solved if the drivers can be informed in advance about the availability of parking spaces. In later progresses in making low-cost, low-power embedded frameworks are making a difference designer to construct modern applications for Internet of Things. The system helps user to know the availability of parking spaces on a real time basis. Based on that the user knows the availability of free space before starts the journey and plan accordingly. Using the tag and detector, amount can be debited digitally and no need of man power. And using camera, free space is recognized and updated in the created application.

Brief Description of the Drawings
The block diagram explains about the process to create device that utilize the benefits of the state-of-the-art technology in terms of adequate performance/power consumption and at the same time being energy autonomous. For this purpose, Bluetooth mobile application was used that according to its specification can achieve which can be considered as enough for smart city parking infrastructure, where simple smartphone can be used as a gateway and connectivity enabler.A vision-based system was developed as an alternative to the ordinary system to detect available parking bay locations.
Detailed Description of the Invention Fig 1 is a front view of the parking slot
The user finds the parking space by using the mobile application created. Information about parking space is given to user by application. User wants to sign in into the application to know about the details. Which parking area gives access to application, the parking area is linked to the application we created. Then in the application user can see the available area in the slot. The camera fixed in the parking slot is used to load the information in the application. First they sign in into application with the basics information like username, password, date of birth, Address. These data are maintained in the cloud. Whenever the user uses the application they need to login using their Id and password. Then the information can view by the user. The user reaches the area by knowing the free space.
In front of the parking area there is a RFID reader use to read the RFID tag placed in the car. By means of the slot present inside is reduced and it is also used for paying money for parking the car. If the user is not interested they can also pay manually. The RFID reader sends the information to the arduino NANO.
The arduino Nano is used because it is tiny in size and used to store more data than Uno. Uno has 2kB SRAM space but Nano has 8kB SRAM space in the system.

Nano is powered by an Atmega328 processor operating at 16MHZ, 2KB of RAM, has -14 digital 1/0,6 analog inputs, and has 5V and 3.3V power rails. So here we used arduino Nano. The Nano is used to pass the information to LCD for more clearance.
While the car is entering the no of vacant places, entering time are displayed on the LED. And in meanwhile the information is displayed in the application under their user information.At the time of departure of the car the no of vacant' places, entering timing, departure timing are displayed on the LED. And also in meanwhile the information is displayed in the. application under their user information.
The video is captured in the parking slot it is used to pass the information to application. The application is connected to the arduino and the parking slot for updating the information on time. So the user can get the information quicker. These user databases are maintained in the cloud. When entering the parking they can enter in the name of the username in the slot for convenience.
Fig 2 is a block diagram illustrates the process of detection and controlling the slot vacancy.
Input Video Frame
The system identifiescars using videos instead of using electronic, sensors fixed in the parking area. A camera is installed at the front of the parking slot. It will record the video of the parking slot. Suppose the car parking system consists often slots to park the whole view of the slot is covered by the camera and the video is captured. Based on the car entry and depart are captured in the video. Setting image of the car as reference image, the image is compared with input video.The capturedvideo then undergoes image processing techniques.

Preprocessing
Initially, the video does not include any car. An empty image of the parking slot is recorded, then the parking slot with cars is also recorded and subtracted. The video' captured is preprocessed for removing noise for further processing and increases the detection process more clear. Background subtraction is the first step in preprocessing used to make the video accurate for image detection and further process. After preprocessing the image is segmented into section. This section is used to increases the chances of redundancy of information.
Feature Extraction
The preprocessed image is then used for feature extraction. In feature extraction the extracted image is compared with the history of frames that stored in computer.The Haar-like features are typically used to determine the information of a region rather than the raw pixel values. From the utilization of a group of Haar-like features, the determination of features and computed values are used as input into a decision tree classifier for identification. On high-end elaboration algorithms such as deep learning, to acquire images of the parking area and recognize the presence or absence of vehicles easily. Here the Region based Convolution Neural Network (R+CNN) is used to find the parking occupancy and smart parking system. R+CNN is used to localize the image. It consists of bounded boxes and detect if any of these blocks contains object. With the comparison of image stored in dataset and recognized image the free slot is identified.
A "region proposal system" or Regions + CNN (R-CNN), where after the final convolutional layers, a regression layer is added to get a number that consists of four variables xO, yO, width, and height of the image.The kernel size remains fixed for both convolution layers and is 5><5 where the size of the pooling area is 2x2 in both subsampling layer.The images of the input are 32x32 that are considered as 1024 linear nodes on which convolution process is to be accomplished. Convolution operation with kernel spatial dimension 5 converts 32 spatial dimensions to 28 (32-5+1) spatial

dimension where the size of the image is 32x32 and the size of the kernel is 5*5. Hence, the convolutional layer (CI) with a kernel size of 5x5 and 16 kernels gives an output of 28x28 in first convolution.
The max pooling procedure is used with the size of 2x2 and we used ReLU as an activation function. The feature maps of size 28x28 are subsampled by a 2x2 window and 16 feature maps with size 14x14 are achieved in subsampling layer SI. The output data from S2 is convoluted with 32 filters of size 5x5 in convolution layer C2. The output data in C3 are 32 feature maps of size 10x10. The output data from past layer is subsampled by a 2x2 window to generate 32 new feature maps with a size of 5x5 in subsampling layer S2. Lastly", hidden layer nodes are connected to the 2 output layer neurons to classify images into 2 classes.
Classification and occupancy output
The basis of object identification in this paper was based on the utilization of Haar-like features. The tool for object classification was developed in an open source library called Open Computer Vision Library (OpenCV). In order to train the object classification algorithm, two sets of images are required to train the classifier. One set of input images would be images that contain the object to be detected, which can be called as positive images, and another set of images that do not contain the object, that are called negative images. In the training of the classifier, the location of the object within the positive image including the height and width of the object need to be specified.
Based on the extracted images the slots are classified as free slot and occupied slot.lt detects the free and occupied space in parking slot based on the image of a car as reference image, the captured images are sequentially matched using image matching. The slot information is transferred to Microcontroller through USB. Then empty slot information is updated in developed application.