Full Project – Motion detection using OpenCV, webcam and Python Flask framework
Click here to Get this Complete Project Chapter 1-5
CHAPTER ONE
INTRODUCTION
1.0 BACKGROUND OF THE STUDY
Intelligent transportation system provides an attractive alternative to the traditional traffic systems, which depend almost on the facilities of the system for traffic circulation and safety. A video camera, coupled with computer vision techniques, makes up of a video-based intelligent transportation system (Zhou et al, 2 007). Detection of moving objects is the first relevant step in this system. To meet the special requirements of efficiency and accuracy of a successful video-based system, moving objects detection algorithm should be characterized by some important features, such as accuracy, real-timeless, etc.
The accuracy of detection is a basic requirement of the system. In general, the accurate detection is time-consuming. Moreover, a real time system ensures that the detection information is provided in time, and the management commands from the control center are responded timely. In fact, a precise moving object detection method makes tracking more reliable and faster, and supports correct classification, which is quite important for a system to be successful (Cucchiara et al, 2003).
During the past decades, researchers in vision technique have already proposed various algorithms for detecting moving objects, such as, consecutive temporal difference(Consecutive frames subtraction) (Fu-Yuan et al, 2005), optical flow approach (Kinoshita et al, 2006) and background subtraction etc. Among these methods, background subtraction algorithms are most popular, because they are relatively simple in computing in a static scene. However, the background is assumed to be static in this method. Thus, shaking cameras, waving trees, lighting changes are quite probable to cause serious problems to a background subtraction model (Roth, 2005). In addition, a successful background subtraction method needs to model the background as accurate as possible, and to adapt quickly to the changes in the background. These requirements add extra complexity to the computation of the model and make a real-time detection difficult to achieve. Optical flow approach is quite excellent because it can detect the moving objects independently and it works very well in changing environments, even in the absence of any previous information of the background.
However, the computational cost of the approach is very expensive, which makes it very difficult to be applied in a real-time system. And, this approach is quite vulnerable to disturbs, such as the headlights of the vehicles. Thus, it is not fit for the traffic control system.
Temporal difference is the simplest method to extract moving objects and robust to dynamic environments. However, it easy to cause small holes and cannot detect the entire shape of a moving object with uniform intensity. Also, any changing elements in the background can be easily classified as the foreground by temporal difference. In this paper, we propose an improved temporal difference approach. Three consecutive frames are used for computing. The pre-treatments are applied to compensate the camera motion in the input frame. Moreover, the post-treatments are employed to optimize the differential results by filling the small holes and removing uninteresting moving objects. The rest of the paper is organized as follows. Section II presents the main results of the paper. First, an overview of the framework of the proposed method is described. Then, the details of the implementation are presented, including camera motion compensation, improved consecutive temporal difference, small holes fulfillment and uninteresting motions removal. The efficiency and accuracy of theproposed method are illustrated by some experiment tests in Section III. Finally, Section IV concludes the paper with some remarks.
The process of image processing is mainly depends on the process of “illuminations” from source and “reflectance” from the object or some absorbtion from scene we can acquire an image. In modern technology motion detection plays a vital role in the field of real time application for an artificial intelligence. Motion detection process is used to find out the displaced object from a relative scene. The motion we get from that process may be a salient one, or it may be distractive one. More number of approaches are followed to find out displaced object through motion detection on a real time video streams. The displaced object from a scene is identified by the help of optical flow vector method. The motion flow is used for monitoring the 2D environment while the advanced 3D technology is process with the help of optical flow field. The optical flow algorithm will make several number of assumption before getting in to calculation. Changes in the illumination of a respective scene and surface reflectance will cause some violation to our assumptions. Inconsistencies in the optical flow field are possible through occlusion effect. The best choice for the optical flow method is Lucas-Kanade algorithms. This algorithm works by comparing the two successive image frame based on that it estimates the displaced object from that two successive image scene. The moved object from a scene is highlighted by the optical flow vector. Lucas-kanade algorithm doesn’t need to scan the next image for matching the pixel of image or neighborhood pixel.New innovative technology revolves around how much a product is capable of implementing along with its price. The Raspberry Pi crosses off both criteria becauseit is a cheap effective computer which is capable of much more.
What makes it soconvenient is that so much can be done with it from a security system to a VPN server. The possibilities are endless! Like any other computer it can accept several programming languages including Python. Most importantly, security can be a necessity today and the Pi has the ability to become a camera security system with a cost under 80 dollars. Regular security systems lead up to prices within the range of thousands. Who would want to buy a single camera for over 100 dollars just to setup on their front door, when they can buy a 29 dollar camera which even notifies them via email? You would never have to worry about looking back through recordings because the Pi Security System would send you an email whenever someone comes by your home. Most of all the Pi Security Camera system is very user friendly. Anyone who has the required materials can do it with a few additional installations of files and save themselves a great amount of money. Not to mention, they would gain an efficient security system.
1.2 Statement of Problem
Security of lives and properties has been of great interest for universe since the exemption of 21st Century. Insurgency attach to civil right of the federalism has caused destructive event in country and world at large. This problem has lingered for generation. Physical deployment of security personnel has disappointed the entire human over decades due to short falls of human’s i.e inaccuracy, incompleteness, lost of integrity and compromise etc. manually implemented security strategy are sometimes inaccurate and are real-time delivery of reports about an event. Timeliness to operation is requirement to a proper security. The problems of the existing system;
- Inaccuracy in object detection
- Detect image in non-real time
- Human eye meant not capture object as fast as possible
1.3 Aim and Objectives
This research work aimed at developing software application, a web based application housed by a browser programmed using a framework called flask python. The program should be able to detect an object on motion. The objectives are as fellows
- Python application that manipulate system device and properties
- A web based platform to detect the webcam
- A platform with capability of capture frame in mjpeg motion format
- A platform to detect object on light/motion
- And draw a rectangular box on the object.
- A system to perform a kind of surveillance operation
1.4 Significance of the Study
To build a system that can solve the above mention problems at the statement of problems. To implement a system capable provide all the objectives. The importance of the system is that in the absence of human security a machine or a program can provide surveillance i.e artificial intelligence
1.5 Scope and Limitations
The research work is implemented with python flask framework with additional software and device i.e Opencv and System Webcam. At its present state is locally hosted. But this system could also be executed on linux ( Ubuntu 10.x) , Windows and Mac Operating System.On different client not more than 20 personal computers exceeding that the server will have difficulty in handling the request.
The Limitations of this research work is that the software cannot be executed in a mobile device but can be view in the android browsers. The following are the shortcoming of the project work, this includes;
- Project given duration, this time range give is very poor to achieve the mean target of this research work.
- Financial implication is great challenge to every student who self-sponsor in the educational career.
1.6 Definition of Terms
Opencv: (Open Source Computer Vision) is a library of programming functions mainly aimed at real-time computer vision used for direct linking with the webcam hardware.
Webcam: is a hardware video camera that feeds or streams its image in real time to or through a computer to computer network and usually located at the front of the screen.
Operating Systems: is system software that manages computer hardware and software resources and provides common services for computer programs and a program that control the entire functionality of the system.
Motion Detector: is a program link to a device that detects moving objects, particularly people or objects and this device is often integrated as a component of a system that automatically performs a task or alerts a user of motion in an area.
Get the Complete Project
This is a premium project material and the complete research project plus questionnaires and references can be gotten at an affordable rate of N3,000 for Nigerian clients and $8 for international clients.
Click here to Get this Complete Project Chapter 1-5
You can also check other Research Project here:
- Accounting Research Project
- Adult Education
- Agricultural Science
- Banking & Finance
- Biblical Theology & CRS
- Biblical Theology and CRS
- Biology Education
- Business Administration
- Computer Engineering Project
- Computer Science 2
- Criminology Research Project
- Early Childhood Education
- Economic Education
- Education Research Project
- Educational Administration and Planning Research Project
- English
- English Education
- Entrepreneurship
- Environmental Sciences Research Project
- Guidance and Counselling Research Project
- History Education
- Human Kinetics and Health Education
- Management
- Maritime and Transportation
- Marketing
- Marketing Research Project 2
- Mass Communication
- Mathematics Education
- Medical Biochemistry Project
- Organizational Behaviour
- Political Science
- Psychology
- Public Administration
- Public Health Research Project
- More Research Project
- Transportation Management
- Nursing
Full Project – Motion detection using OpenCV, webcam and Python Flask framework