Save The Drowsy Driver Drowsy Driver Detection

DOI : 10.17577/IJERTCONV9IS07013

Download Full-Text PDF Cite this Publication

Text Only Version

Save The Drowsy Driver Drowsy Driver Detection

Adithya Sajikumar

Dept. of Computer Science and Engineering Mangalam College of Engineering Ettumanoor, India

Aravind Aji

Dept. of Computer Science and Engineering Mangalam College of Engineering Ettumanoor, India

Anandu Prasad

Dept. of Computer Science and Engineering Mangalam College of Engineering Ettumanoor, India

Arun K Stanly

Dept. of Computer Science and Engineering Mangalam College of Engineering Ettumanoor, India

Jibin Baby Neethu Maria John

Dept. of Computer Science and Engineering Asst. Prof, Mangalam College of Engineering Mangalam College of Engineering Ettumanoor, India

Ettumanoor, India

Abstract – AI procedures are these days to anticipate the state of a driver to give data which can improve wellbeing out and about. It is a use of man-made consciousness. frameworks can naturally learn just as improve without being expressly modified utilizing Artificial Intelligence. A driver's condition can be assessed by bio-pointers utilizing monochrome cameras while driving just as the appearances on the substance of a driver. In this paper we present a comprehensive and complete study of ongoing works identified with driver languor recognition and ready framework. We additionally present the different AI procedures like PERCLOS calculation, HAAR based course classifier, OpenCV which can be utilized to decide the driver's condition. At last, we distinguish the shortcomings looked by the flow arrangements and present our examination results.

Keywords – Artificial Intelligence, Autonomous Vehicle Technology, Drowsiness Detection, Machine Learning.

  1. INTRODUCTION

    Resting is one of the essential requirements of the person and rest need makes the body become wastefully, lessening both response time and energy, additionally produce low readiness and absence of fixation which decreases the capacity to play out a few exercises dependent on care that is fundamental on account of driving a vehicle. Sleepless driving is a significant reason for vehicular mishaps. As indicated by numerous explores sleepiness is identified with a great many car crashes every year, the mishaps create roughly half of death or genuine wounds, as they will in general be impacts at high velocity in light of the fact that the driver who has nodded off can't break or go astray to stay away from or decrease sway. To alleviate these mishaps, makers have created sleepiness discovery frameworks that perceive indications of conceivable tiredness, making the driver aware of their condition. This framework is created as an application for an Android-based cell phone, where estimating security-related information that doesn't need extra expenses.

    The system has an efficiency of 96% to detect that the driver is awake and 97% to detect that he is asleep. This information allows knowing the signs that shows a sleepy driver. The following contains a concise description of the papers we studied. The paper presents an arithmetic based method to solve the problem related to the detection of drowsiness. Three stages are involved here. They are Face detection, Eye position detection and Eye tracking. This paper provides an efficient method for the detection of the state of the driver. This framework uses the motion of the eyes to detect the state of the driver and gives an alert within

    0.5 seconds. The performance of the driver is transcribed in the form of a graph. A new method for fatigue detection is presented. YCbCr color space and canny edge detection methods are used here. These methods help to determine if the driver is under fatigue. When the driver is drowsy or sleepy, an alarm system is turned on. A distinct system which utilizes the concept of computer vision and imaging is designed. A software algorithm is developed. This algorithm is partially tested and is found to be working effectively. Research is in progress in order to develop a full-blown system. The developed system is capable in identifying the state of drowsiness quickly. The system is capable of differentiating between normal eye blink and the eye blink associated with drowsiness. It is capable of performing under low light conditions and when the driver is wearing spectacles. This can further be developed by adding different sensors. The developed system is based on computer vision. The system utilizes Viola Jones algorithm as well as the CAMSHIFT algorithm. This paper is concerned with the development of a software framework for the timely and precise detection of drowsiness. Multiple facial features were considered as inputs.

    It depends on the idea of picture preparing. The framework gives a non-intrusive methodology. This framework likewise proposes the joining of yawning as a boundary to identify sluggishness. To assess a driver's condition, certain facial highlights were distinguished. Utilizing python libraries, they were inspected. These highlights were pace of eye conclusion, ECD, per conclusion, head positions and pace of yawning. Certain limits were likewise featured.

    Fig: Use case diagram

  2. PROPOSED SOLUTION TO DETECT DROWSINESS

    This part portrays various parts of the framework considered in its execution; they incorporate the practical necessities just as the instruments utilized and gadgets chose for framework testing in various investigation cases. The pre- owned calculation measures the shading data present in the picture, changing it over to grayscale. To decide the face in the picture, the picture is isolated in sub areas deciding if the subregion is a face or not. The utilization of this calculation implies an efficient and just the subregions that contains a face are prepared. The motion location is done from the lingering mistake that is demonstrated considering a direct mix of facial development models. A comparable model is considered to distinguish the position and tendency of the face. It incorporates a framework that permits recognizing facial motions within the sight of head development. Figure shows the flowchart of the framework. The code used to carry out the calculations was made considering the restrictions that have cell phones like the restricted substance or highlights in the interface, moderate or restricted equipment and use circumstances. Their achievement in usefulness depends in transit they are planned and advanced by organization that possesses Android. The application is introduced on driver's gadget running Android working framework (OS). The cycle begins with catching of live pictures from camera and is

    therefore sent at nearby worker. At the worker's side, Dlib library is utilized to recognize facial tourist spots and a limit esteem is utilized to identify whether driver is sleepy or not (T. Soukupova and J. Cech,2016). These facial milestones are then used to process the EAR (Eye Aspect Ratio) and are returned back to the driver. In our specific circumstance, the EAR esteem got at the application's end would be contrasted and the edge esteem taken as 0.25(T. Soukupova and J. Cech,2016). In the event that the EAR esteem is not exactly the limit esteem, this would show a condition of weakness. If there should be an occurrence of Drowsiness, the driver and the travelers would be cautioned by an alert. The ensuing area subtleties the working of every module.

    Fig: System Architecture

    Fig: Flowchart

  3. RELATED WORKS

    The created framework is a constant framework. It utilizes picture preparing for eye and face discovery. HAAR based course classifier is utilized for face identification. A calculation to follow objects is utilized to follow the eyes persistently. To recognize the tired condition of the driver, the PERCLOS calculation gave [2]. The paper centers around building up a non-nosy framework which can distinguish weariness and issue an admonition on schedule. The framework will screen the driver's

    eyes utilizing a camera. By building up a calculation, the side effects of driver weakness can be recognized early enough to keep away from mishap. At the

    point when the indications of exhaustion have been distinguished yield as sound and safety belt vibration is given to alarm the driver. Cautioning will be deactivated physically as opposed to consequently. This paper utilizes a quicker calculation than PERCLOS. This framework will recognize driver's weariness by the preparing of the eye area. After picture obtaining, the principal phase of preparing is face identification. If the eyes are blinking without any delay and if the eyes closed more than 0.5 seconds, this system issues warning to the driver. The warning is in form of an alarm and vibration. MATLAB is used for the processing of the image. System makes use of the number of eye blinks for detecting the state of drowsiness in a driver. The system makes use of Open CV and Raspberry Pi module with a single camera view. The eye status is obtained through image processing algorithms. This paper takes into account only the state of the eyes, it does not focus on the frequency of yawning. In this system computer vision and alcohol gas sensor combination is used to detect drowsiness and alcohol intoxication. This system makes use of Raspberry-pi and Arduino UNO with.mThe proposed system is based on computer vision and embedded system applications. Eye closure is detected using HAAR based cascade classifier and an alcohol gas sensor which functions as a Breathalyzer. This system includes two modules. The two modules are the face and eye detection module followed by the face tracking module. CAMSHIFT algorithm is used for continuous face tracking. This system also uses cascade classifiers

    in order to improve the accuracy of face detection. The system is a real time non- intrusive model. To reduce the number of accidents caused by drowsiness, various methods for detecting drowsiness automatically have been developed. Three ideas are discussed in this paper; the first idea is creating a dataset of drowsy facial expressions. The second idea is to combine visual, non-visual, and vehicular features into one. The last idea is to develop wearable hardware such as smart watches in order to detect

    drowsiness. The framework will screen the driver's eyes utilizing a camera. By building up a calculation, the side effects of driver weakness can be recognized early enough to keep away from mishap.

  4. METHODOLOGY

    After surveying a number of different papers, the following methodologies have been identified:

    A. PERCLOS

    Initially, in order to identify the drivers drowsy state using PERCLOS, we need to perform the following steps as per [2]:

    • Perception of face and face pursuit.

    • Position of eye and eye pursuit.

    • Identification of the state of the eyes.

    • Calculation of percentage of eyelid closure.

    • Identification of the drowsy state.

    PERCLOS is one of the measures to notice the wstate of drowsiness. The PERCLOS (Percent of the time Eyelids are shut) measurements is utilized to measure sluggishness in the work "Eye following based driver weakness observing and cautioning framework" [9]. The framework gauges with a non – parametric techniques for recognizing laziness, the vehicle directing wheel changeability is considered to decide the measure of sleepiness since drivers makes inconstancy more noteworthy as driver become more sluggish. The PERCLOS measurements for cautioning driver is utilized in [10] to identify tiredness in hefty vehicles, to screen and caution the driver [11], for line takeoff admonitions [12] and to identify languor conditions in drivers [13].

    1. CAMSHIFT

      A robust and nonparametric technique is used [2]. It implements the CAMSHIFT algorithm. CAMSHIFT (continuously adaptive mean-shift) is an efficient and light weight tracking algorithm. It is based on the concept of mean shift It is suitable for tracking targets in simple cases [2]. It is not efficient in tracking objects in complex situations. A detection algorithm can be applied to successive frames of a video sequence to track a single target. According to the detection algorithm [2] can be described by the following steps:

      1. Initialize the size as well as the position of the search window.

      2. Calculate the mass Centre (Xc, Yc) of the window.

      3. Adjust Centre of the window to mass Centre.

      4. Repeat 2 and 3 until distance of the two centers (Centre of the window and the mass Centre) is less than some threshold value.

    2. HAAR TRAINING

      The Open CV library gives various capacities to face and highlight (eyes, mouth, shades, and soon) recognition. A portion of these capacities can be utilized to prepare classifiers. The classifiers can be prepared for the cycle of recognition of face This is known as HAAR preparing. Item. Here, a course work is prepared from various pictures, both positive and negative. Each component is a solitary worth gotten by taking away amount of pixels under different areas of the pictures [3]. The pixels utilized for extraction is distinctive for each component. Every one of the extricated highlights will not be helpful for the necessary cycle.

    3. VIOLA JONES ALGORITHM

    Viola Jones algorithm uses the following techniques in its algorithm [8]. They are:

    • HAAR based features

    • Integral Image Formation

    • AdaBoost Technology

    • A cascade of classifiers

    Highlights are chosen dependent on the pixel powers in HAAR based component portrayal. It doesn't mull over, the upsides of the pixel. HAAR based highlights are scalar item between the picture and some HAAR layouts. Essential picture development is utilized for highlight estimation. It thinks about just four corners of the picture. Versatile boosting (AdaBoost) is utilized to choose the necessary highlights. Because of the utilization of Adaptive Boosting there is a decrease in the computational season of the calculation. A course of classifiers is utilized to build up a solid classifier chain. The OpenCV library gives an order brief preparing utility called HAAR- preparing which produces a classifier in XML design when given positive and negative instances of the item to be distinguished.

  5. RESULT

    Using this system, we can ensure an efficient system which can be used prevent accidents due to drowsiness to a large extend. This system can be implemented in all kinds of car. We can also add additional implements which can include a message sent to another person which can be pre-selected by the user, and the message can include details like GPS location, driving speed etc. at the point of a drowsiness detection. All these can be combined so that a large number of accidents can be reduced and a lot of people can be saved from severe injuries and death.

  6. CONCLUSION

    This paper gives a similar report on papers identified with driver laziness recognition and ready framework. To give an answer for the issue of recognizing the condition of languor, a math based strategy is utilized [2]. This framework utilizes eye development to identify exhaustion. Eye development is distinguished utilizing a camera. This is done to perceive the indications of exhaustion to stay away from mishaps [6]. It depends on the idea of eye-following. To acquire better outcomes, hundred and fifty pictures of various individuals have been utilized. In the event that the condition of weakness has been recognized, a caution framework is turned on [9]. PC vision with installed frameworks ar utilized. A product calculation is created. It was halfway tried and discovered to be compelling. There is a lot of extension for additional upgrades [4]. The proposed framework recognizes laziness if the eyes have been shut for a time of at least four edges. The recognition framework separates the ordinary eye flicker from sleepiness. The created framework is a non-obtrusive framework. The framework can be additionally evolved by adding different sorts of sensors [5].

  7. REFERENCES

  1. Kyong Hee Lee, Whui Kim, Hyun Kyun Choi, Byung Tae Jan. A Study on Feature Extraction Methods Used to Estimate a Drivers Level of Drowsiness, IEEE, February 2019.

  2. Lorraine Saju, Christeena J, Farhana Yasmin, Surekha Mariam, Drowsiness detection system for drivers using HAART training and template matching, IJEAST, Vol. 1, Issue 6, April 2016.

  3. Dwipjoy Sarkar, Atanu C, Real Time Embedded System Application for Driver Drowsiness and Alcoholic Intoxication Detection, IJETT, Volume 10 Number 9, April 2014.

  4. SrinivasuBatchu, S Praveen Kumar, Driver Drowsiness Detection to Reduce the Major Road Accidents in Automotive Vehicles, IRJET, Volume 02 Issue 01, April 2015.

  5. Hardeep Singh, J S Bhatia and Jasbir Kaur, Eye Tracking based Driver Fatigue Monitoring and Warning System, IEEE, January 2011.

  6. Fouzia, Roopalakshmi R, Jayantkumar A Rathod, Ashwitha S, Supriya K, Driver Drowsiness Detection System Based on Visual Features. , IEEE, April 2018.

  7. Varsha E Dahiphale, Satyanarayana R, A Real-Time Computer Vision System for Continuous Face Detection and Tracking, IJCA, Volume 122 Number 18, July 2015.

  8. SaeidFazli, Parisa Esfehani, Tracking Eye State for Fatigue Detection, ICACEE, November 2012. Gao Zhenhai, Le DinhDat, Hu Hongyu, Yu Ziwen, Wu Xinyu, Driver Drowsiness Detection Based on Time Series Analysis of Steering Wheel Angular Velocity, IEEE, January 2017.

  9. Bagus G. Pratama, IgiArdiyanto, Teguh B. Adji, A Review on Driver Drowsiness Based on Image, Bio-Signal, and Driver Behavior, IEEE, July 2017.

Leave a Reply