Facial recognition involves identifying human beings based on the distinct features of their faces. Any individual who has been tagged an image on Facebook or any other social media site are familiar with facial recognition technology. Facial identification technology has evolved beyond finding friends in social media; presently, it has been used to track people as they go to church, travel and even shop. Several government agencies such as Department of Homeland Security, CIA, FBI, and DEA are using this type of technology to ensure the borders are secured. Also, facial recognition technology can be used by government agencies to detect and catch drugs smugglers. Moreover, facial recognition technology Also, Casino uses facial recognition technology to track their high-rollers and keep out card cheats. Lastly, retailers can use Facial recognition technology to get to know who shop in their stores and catch shoplifters.
Facial recognition technology depends on software that is designed to images of people’s face from vying to photos and identifying individuals by matching the pictures of individuals with those stored in the database (Raghavendra, Raja, Kiran, Busch, Christoph, 2016). Facial recognition technology is cheap to implement because it can be used where there is a camera or CCTV surveillance. Cameras and Facial recognition technology can be incorporated together to ensure security is to enhance because many cities have cameras installed. Most companies that develop facial recognition technology have ensured that their software is efficient and reliable. Most of the facial recognition software can process close to one million facial matches in a second per server.
Delegate your assignment to our experts and they will do the rest.
The growth of facial recognition technology has outpaced efforts to regulate it, and it has caused some discomfort among private lawyers. America, for instance, is the leading nation to embrace facial recognition technologies since most states have no rules governing the collection and use of data gathered biometrically (Blackburn, Bone, and Phillips, 2001). Only two states that are Texas and Illinois have laws that regulate collection and analysis of biometric data. However, the law was passed before the facial recognition technology had emerged. Several states have acknowledged and accepted that facial recognition technologies are the way to go to enhance security, reduce theft and eradicate selling of prohibited drugs.
Facial recognition permits tracking of individuals secretly when they are in the public domain. The facial recognition tracks individuals whether a person is going to church, bar or visiting a doctor ( Achterberg, Trezza, Siviy, Schrama, Schoffelmeer, Vanderschuren 2014) . The technology is crucial because it monitors the individual’s movement of a person. However, some legislatures and advocates have raised concerns over facial recognition technology because it intrudes into people’s privacy which is a right and it is enshrined in the constitution. The opponents of facial recognition technology argue that people’s privacy should be upheld even if they are in public.
However, a system that automates facial recognition technology is promising but not for many law enforcers. When an organization such as an airport, and business consider employing automated facial recognition technology as an application the question that the decision makers in the organization should ask themselves is Which system works best?”. This question has no direct and simple answers because there is no single performance rational that that could be used when comparing the facial recognition technology.
There are many types of facial recognition technologies that can be used for drug screening purposes. Some developers of facial recognition would quote performance to organizations software such as false reject rate (FRR) or false acceptance (or alarm) rate (FAR) when recommending the best software. FRR is the percentage of valid users wrongly rejected by the system whereas FAR is the percentage of imposters wrongly accepted by the system. The limitation of using these statics include 1) Both FRR and FAR are dependent upon the precision of the application such as user motivation; 2) There is a tradeoff between FRR and FAR which depends on the throughput requirements and security of the system.
Also, some sellers can quote an equal error rate (EER) in the environment of the laboratory. ERR occurs when FRR is equal to FAR. However, ERR does not depict the entire story since it depicts a performance measure for a single throughput/security requirement in an environment that is single (Osselton, 2017). When evaluating the facial recognition software, it is important to consider error rates over a range of environment or settings.
Theoretically, verification involves one to one comparison where a user avails an identity along with the biometric system and a live biometric determines whether the live template that is acquired by the sensor is similar to those stored under the claimed identity (Osselton, 2017). The biometric is identification is one too many where the image of the live biometric is used to be compared to the images stored in the database in verification applications errors depends solely on FRR and FAR setting. In application used for identification, error rates depend on the magnitude of the searched database and FAR/FRR settings.
In this paper, we will depict for the first time that facial recognition technology can be able to detect a different kind of drugs such as cocaine, opiates, and amphetamine through facial scanning (Osselton, 2017). Drugs such as cocaine, opiates, and amphetamine affect people in the same way. So, different drugs have different ways to bring out facial expression. The presence of drugs in individuals was confirmed using FRVT 2000 test design and Drug Abuse Database (DAD) by analyzing different facial expression in the database with the face of an individual suspected to be under the influence of drugs. The result highlights that a person can screen for drugs using their face.
Method
Drug Abuse Database (DAD) can provide information about drug addicts basing on the analysis of different facial expression that drug users exhibit when under the influence of certain drugs such as cocaine, opiates, and amphetamine. Fase expression was collected from individuals who were under the influence of cocaine, opiates, and amphetamine using cameras and pictures.
Sample collection and ethics
Staffs of rehab collected each single face sample by taking facial pictures of individuals who were under the influence of cocaine, opiates, and amphetamine. Individuals who are brought to rehab and are high on cocaine, opiates, and amphetamine are photographed and recorded by the cameras. The time frame is important because when the photography and recording are not conducted immediately cocaine, opiates, and amphetamine which is in the system of the drug user start to reduce and it will be difficult to capture the facial expression of individuals who are under the influence of the three drugs. Once the sample is collected analysis is done using FRVT 2000 is a test design. The potential concern when detecting drugs using face expression is that individuals may fake the expression giving wrong information. Individuals who participate in this should be coerced to relax so that the true facial expression are revealed.
The process for analysis of the facial sample collected is as follows. The picture captured is printed to provide a physical comparison. The facial recording from the cameras is compared against the DAD who has standard images that depict the facial expression when a person is on cocaine, opiates, and amphetamine (Osselton, 2017). The analysis of the images is displayed on the screen after ten minutes of initializing the test. The result of each test gives either match or no match from the presence of drugs that causes people to make different facial expressions.
The facial expression of individuals under the influence of cocaine, opiates, and amphetamine was collected in three United States of America States which were involved in the study. These three states are Florida, Illinois, and New York ( Acosta, Eissenberg, Nichter, Nichter, Balster 2008) . Under the laws of each state, facial features samples were collected from the routine photography and recording of new addicts brought to rehab. Rehab workers were asked not to preselect cases to ensure there is variance in the data they collected and a full series of continuity of evidence were maintained during the sample analysis and collections — the time of being intoxicated by cocaine, opiates and amphetamine and sample collection speckled between cases.
Each sample of the faces was analyzed for the following drug groups: amphetamine, opiates, and cocaine. Two sets of analyte data were gathered from each: 1) results of quantitative tests conducted by the independent third-party analytical laboratory on face expression sample gathered in parallel with the facial testing ( Attwood, Penton-Voak, Munafó, 2008) Results of matching tests conducted within the rehab center by using FRVT 2000. The result of these two tests was used to evaluate the accuracy of facial recognition technology and FRVT 2000: the data and the test results which were collected anonymized.
Confirmatory analysis
As specified above for confirmatory examination of the presence of drugs in a person through facial features and expression a second image and video are recorded and captured from each. The images and video were collected by a camera by rehab staffs. The sample is properly packed and sent to a laboratory to be analyzed. The confirmation analysis of the drugs present in an individual is done through FRVT 2000.
Results and Discussion
The facial identification application uses a three-phase assessment protocol recommended in “An Introduction to Evaluating Biometric Systems (Osselton, 2017).” The FRVT 2000 technology results and set-up assessments determine the performance level that can be projected by utilizing the tested methods underneath the applied testing standards. The study applies the results of real analysis performance in specific applications. The images displayed in the resultant scenarios were obtained from the FRVT 2000 Evaluation Report ( Attwood, Penton-Voak, Munafó, 2008). As from the report, the results obtained here should be a reference point rather than exclusively the others displayed below when conducting other case studies. Comprehensive information illustrating all graph depicted below is as well provided in the FRVT 2000 Evaluation Report.
In the case of persons obtaining traveling documentations are compared to those disembarking the passage, the suppose there is a transportation mode in which people register to embark then disembark in later hours at some other location (Raghavendra, Raja, Kiran, Busch, Christoph, 2016). We may assume that every person disembarking is one person registered for a specified boarding pass. The imaging procedure requires cooperation from the passengers in both registration and disembarkation. The case distinguishes imposters 99% of the time. Thus the FAR is assumed to be at 1 percent (Osselton, 2017). Having all these considerations in one place, the FRVT 2000 technology verification results can be used in the evaluation of the projected performance.
Compression: The compression effects relates to the performance factor. Nevertheless, verification compression scores tests were not included in FRVT 2000.
Distance: the distance is not the major concern as it should be practicable to make sure that the camera-subject distance during disembarking is the equivalent to that applied for registering.
Expression: Facial expressions of passengers can differ at both registration as well as disembarking time.
Media: The assumption, in this case, is that similar media is applied to capture registration and disembarkation images ( Achterberg, Trezza, Siviy, Schrama, Schoffelmeer, Vanderschuren 2014) . Therefore the media is not a factor of performance factor in this case.
Pose: The inclination of the pose have to be the same for registering as well as disembarking. Thus pose is excluded as a performance aspect in this scenario.
Resolution: In line with the distance analysis resolution should be factored out.
Temporal: The time-based effect is not relevant in this scenario as registration and disembarkation take place with insignificant time variances.
If the assumption is that performance is suitable, the “Enrollment Timed Test” results should be considered of which is identical to this case (assumption of similar camera system as well as lighting) ( Attwood, Penton-Voak, Munafó, 2008). In the authentication segment of this scenario assessment, both vendors succeeded to match subjects in every single trial except for one (whereby the two systems are crashed leading to, no probe images was acquired) utilizing 135 gallery images with no forged matches (Osselton, 2017). If these confirm to be suitable performance, the recommendation of the authors is to adopt a more application specific case assessment of performance adhering to operational assessment on candidate configuration.
This paper presents the Drug Abuse Database and Facial Recognition Vendor Test 2000 results fragment, and further, it offers discussion on the application of facial recognition in future relevance to the supply-side drug restrained community. It has been depicted that, at the modern art state facial recognition technologies can execute various errands indispensable to the community with the little help from a human in making final correspondence to the top candidate's list or resolution of proof oversights. Selecting the facial identification structure to utilize for each application is a huge hurdle. People commissioned with the investigation of facial recognition for particular applications should stick to these three-phase process (technological assessment, the evaluation of scenario, as well as operational assessment) that was demonstrated in the result discussion section of this paper.
Reference
Achterberg EJ, Trezza V, Siviy SM, Schrama L, Schoffelmeer AN, Vanderschuren LJ.( 2014) Amphetamine and cocaine suppress social play behavior in rats through distinct mechanisms. Psychopharmacology (Berl); 231:1503–1515.
Acosta MC, Eissenberg T, Nichter M, Nichter M, Balster RL. (2008). Characterizing early cigarette use episodes in novice smokers. Addict Behav. 2008; 33:106–121.
Attwood AS, Penton-Voak IS, Goodwin C, Munafó MR. Effects of acute nicotine and alcohol on the rating of attractiveness in social smokers and alcohol drinkers. Drug Alcohol Depend. 2012; 125:43–48.
Attwood AS, Penton-Voak IS, Munafó MR. Effects of acute nicotine administration on ratings of attractiveness of facial cues. Nicotine Tob Res. 2009b; 11:44–48.
D. Blackburn, M. Bone, and P. J. Phillips. (2001) “Facial Recognition Vendor Test 2000 Evaluation Report.”. Available online at http://www.dodcounterdrug.com/facialrecognition /FRVT2000/documents.htm .
Osselton, M.D. Urinalysis: the detection of common drugs in urine. In: Wolff, K. (ed). Detection of Drug Misuse - Biomarkers, Analytical Advances, and Interpretation. Royal Society of Chemistry: Cambridge, U.K, 2017; pp. 3–22.
Raghavendra, R, B. Raja, Kiran, Busch, Christoph. (2016). Impact of Drug Abuse on Face Recognition Systems
