Today technological advancement has brought about change in science and engineering particularly to aeronautics. In recent years, research has channel towards the use of drones, with significant applications in mines detection, aerial photography, shipping and delivery, geographic mapping, Entertainment, law enforcement, disaster management, precision agriculture, search and rescue, weather forecasting, wildlife monitoring as well as security surveillance. In this work, drone technology was used for security surveillance system with an integration of deep learning approach in object detection and recognition.
Today technological advancement has brought about change in science and engineering particularly to aeronautics. In the last few years, research has channel towards the use of small aerial Vehicles known as UNMANNED AERIAL VEHICLES (UAV) or drones, with significant applications in Military communication, Agriculture, photography, land scaling, for dynamic surveillance systems. Security is a global phenomenon which draws the attention of governments, stakeholders, corporate society as well as individual to make available, a bearable protection mechanism for good, though security as it sound may perhaps not be singleton, it requires several layers of protection to ensure effectiveness. Therefore, the need for surveillance system for detection and monitoring is very essential. Drone is a device that can be used for this scenario, which comprises of four rotors positioned midway to the centre mass of the device in order to produce a lift and manoeuvrabiliby. Basically, electric motor are source of movement for drone with four upward rotors that facilitate any manoeuvre within its flying area. Drone is designed with a closed loop control to maintain a predefined trajectory, also with automatic take-off and landing abilities in some cases and rotates above a particular moving or stationary object of target [1].
In nowadays technology, drone are small in size and use in many applications such as; mines detection, aerial photography, shipping and delivery, geographic mapping, Entertainment, law enforcement, disaster management, precision agriculture, search and rescue, weather forecasting as well as wildlife monitoring shown in Figure 1.

Figure 1: Application of Drone [4]
Top-down view angles and real-time constraints are some of the issues associated with drone application for object detection, monitoring and recognition in video scene for surveillance system.
Related Work
To have a better understanding of drones, the following are some of the existing systems from author(s) point of view.
Jun and Wang [2] proposed an approach that employ drones to monitor and control the emissions of vessels. In their framework, Drone Scheduling Problem (DSP) was model to overcome time-dependent location and multiple trips for drone. The approach of Pulkit et al. [3], employed the use of drone to replicate human presence to war torn areas or carrying reconnaissance missions beyond borders. In their framework, model of the drone is used to study the flight plan of the drone with gravity considerations in non-linear analysis. Bodnar et al. [5], proposed an approach that uses a drone for management of fire in a forest. The usefulness of this approach is to overcome the traditional reconnaissance in an event of fire management. Edwin et al. [6], proposed an approach that use drone for monitoring and controlling smart cities related issues. In their framework, drone is use in the metro city for monitoring cyber security, privacy and public surveillance. The approach of Seon et al. [7] proposed a framework that use a drone for boarder surveillance system. In their approach, drone is use for altitude and geo fencing so as to navigate communication. Guoru et al. [8] employed the use of amateur surveillance system based on cognitive internet of a thing (IoT). In their approach, machine learning is use which reduces human eye effort. Thale et al. [9] proposed an approach that use a drone for home surveillance system. In this work, drone was for object detection and recognition for security surveillance system in a dynamic environment.
Proposed System
In this proposed framework, drone was used for object monitoring and detection in security surveillance system. Flight controller, aviation regulator, cameras and video transmitter, as well as android mobile are the basic technology configured in the drone. The general model of the proposed framework is depicted in Figure 2, where video scene taken from drone was used for object detection and recognition base on trained data.

Figure 2: Overall system
Implementation
The software structure of our system consists of two approach. The first approach is a detection algorithm and the second approach is recognition algorithm which form the basics of deep learning using OpenCV libraries. In this framework, live video streaming from camera was used for the overall surveillance system processes. Firstly, object is detected and monitor via the drone camera and later on recognize the detected object by name classification base on the trained data. However, the result of the whole process are virtualize through an android App.
Figure 3 shown the working principles of the drone motor technology. The motor stator which is the stationary aspect of the rotational system responsible for magnetic field generation during the drone take-off. The drone have four motors which are classified into: Two (2) clockwise (CW) and two (2) counter clockwise (CCW) that are rotated by the motor bell base on the technological configuration. The motor can position itself by yaw (right or left) and also move forward or backward (pitch). Also, the Electronic Speed Control (ESC) is used to monitor and control the speed and the rotational level of the motor direction. The battery configured with solar panel are the main source power to drone motor used in this work.

Figure 3: Overview Design of Motor
Experimental Results
The performance of our approach have been tested practical in field operations and live videos from camera for surveillance in object detection and recognition system. In this paper, result of our approach is been presented in Figure 4.

Figure 4: Proposed Drone
Figure 4 shows the working prototype for object detection, monitoring and recognition in security surveillance system tested in both field operational and videos. The prototype is capable of detecting object via camera mounted on the drone as well as recognizing object (face) base on the trained model. Android Apps is integrated in to the system for live detection and recognition.
Figure 5 shown the result of an object which have been detected via live camera by the prototype. The object was only detected but not recognized because the prototype was not trained on the so-called object.

Figure 5: Detected Object
Figure 6 shown the result of an object that was detected and recognized via live camera. The object was recognized by name because the trained data model on the prototype.

Figure 6: Recognized Object
As security is a global challenge which draws the attention of governments, stakeholders, corporate society as well as individual to make available, a bearable protection mechanism for good, the paper aimed to provide a framework to overcome this challenge through drone technology and the integration of android tools. The approach of deep learning was used for object detection and recognition specifically OpenCV. The overall achievement of this work is to improve the security surveillance system in a dynamic environment.
Roger, C. Understanding the Drone Epidemic. Xamax Consultancy Pty Ltd, Canberra, Australia, 2014.
Jun, X. and K. Wang. Drone Scheduling to Monitor Vessels in Emission Control Areas. Sino-US Global Logistics Institute, China, 2019.
Pulkit, S. and S. Arockia. Conceptual Design and Non-Linear Analysis of Triphibian Drone. Vellore Institute of Technology, Tamil Nadu, India, 2018.
Joshi, N. 10 Stunning Applications of Drone Technology. ttps://www. allerin.com › blog › 10-stunning-applications.
Bodnar, L. et al. Conceptual Approach of Measuring the Professional and Economic Effectiveness of Drone Applications Supporting Forest Fire Management. Ludovika ter 2, Budapest, Hungary, 2017.
Edwin, V. et al. Drone for Smart Cities: Issues in Cyber Security, Privacy, and Public Safety. International Wireless Communication and Mobile Computing Conference (IWCMC), 2016, https://www. semanticscholar.org.
Seon, J.K. and J.L. Gino. Drone Aided Border Surveillance with an Electrification Line Battery Charging System. 2018, https://link.springer.com.
Guoro, D. et al. An Amateur Drone Surveillance System Based on Cognitive Internet of Things. 2018, https://ieeevplore.ieee.org.