Design and Development of Deep Learning Based Drug Distributor Detection System

Abstract

Today toxicity is an important factor in failed drug development, and its efficient identification and prediction is a major challenge in drug discovery. Historically, the deployment of an intrusion detection in a college was fraught with challenges. Intruder and drug dealer detection has gain a broad attention and become a fertile field for several researches, and still being the subject of widespread interest by researchers. The Intruder and drug dealer detection community still confronts difficult problems even after many years of research. Reducing the large number of false alerts during the process of detecting unknown attack patterns remains unresolved problem. The available technologies didn’t provide the flexibility required to do it right. Anomaly detection is a key issue of intrusion detection in which perturbations of normal behaviour indicates a presence of intended or unintended induced attacks, faults, defects and others in schools and colleges. Today, deep learning based intrusion detection systems provide the flexibility and customization required to deploy intrusion detection at a college building. Here, deep convolutional neural networks (CNN) were trained to predict drug distributor or intruder from dataset images. Deep convolutional neural networks are able to classify normal student correctly, and detect abnormal activities without using a huge amount of training data. Experimental results show that our CNN based anomaly detection obtains high accuracy. Moreover, comparisons with other machine learning techniques including KNN, SVM, and Naive Bayes demonstrate that our proposed method outperforms traditional ones.

Existing System

The existing system is a video-based face recognition system using Matlab and Arduino. Capable of identifying or verifying a person from a digital image or a video frame from a video source. This prototype system is used to detect a face, track it, match it with stored Eigen faces and accordingly set digital pin of Arduino board HIGH or LOW. Using MATLAB, face recognition algorithm has been developed with the PCM technique. The Eigen faces are stored first and then we take snapshot of user’s face in real time. Then we match the user’s face with stored faces and we interfaced this Face recognition with Arduino using Serial communication.

Proposed System

In image processing there are four major steps that involves in detecting the human faces in video-based face recognition. It plays an important role in detection of human faces from videos. The steps involved are image acquisition, image pre-processing, segmentation and feature extraction. The first step is image acquisition that is retrieving the facial images from pi camera for identifying the human faces. The image pre-processing is used to convert the original images into the grayscale images or into pixels. During the pre-processing step the noise in the images are removed, so that a good quality image is obtained. The feature from images are extracted and these features helps in detecting the facial characters in human face that is captured from video is obtained. Feature extraction is done using Haar Cascade classifier technique. Haar Cascade classifier is used to detect objects in other images. Finally, detected data are fed into convolutional neural network classifier and the output describes whether there is a drug dealer or a student one in the live video. If it a drug dealer person, then raspberry pi automatically sent SMS to the principal of the University.

Architecture

BLOCK DIAGRAM