1.1 IntroductionAdvertisingis a public and paid announcement of a persuasivemessage by the company. Nowadays,signage had been changing fromtraditional printed signage to digital signage. The purpose of digital signageis to deliver the message and information to an audiencewhich includes news, notice, movietrailer and advertisement. Digital signage can be easily implemented by usingliquid crystal display (LCD) or projection to deliver the message. Digital signage can be easily found almost everywhere such asshopping malls, airports, restaurants, and station.
Digital signage has moreadvantages compared with traditionalsignage because it can display information and advertisement not only in image.It can deliver the message and information using multimedia content as video,animation, and audio. Some digital signage also requiresthe interaction between the audience and the advertisement.Targeted advertising is an advertisement that served as audiencedemographic for the advertiser to know what the target audience.
To establishthe targeted advertising with digital signage, audience biometrics characteristicsuch as the face, iris and gait arerequired for apply biometric detection. In this research, we are usingmultimodal biometric that combine several types of the biometric characteristic for detection. The attributes features that obtain from themultimodal biometric such as gender, age, glasses, and beard will be used as a statistic for a broadcastingselected advertisement for the targeted audience according to audiencedemographics.
Face recognition is a computer application that identifying orverify a person from image and video.Face recognition play an important role intargeted advertising because it can detect the face multimodal such as age,gender, beard and skin color. With theprocessing of the face recognition, it can develop adaptive appearance modelfor monitoring the audience demographic.
1.2 Problem StatementIn order torecognize the audience, face consists the crucial information foridentification. But most of the face detection and face recognition able todetect the frontal view or side view of the face. The result will be affected if putting the side view of theface into the face detection. To avoid this problem, it only requires the frontal view of the face of the audience to classify as the viewerswho look at the digital signage and the side view as the passerby who does not look at the digitalsignage.
Since the real-time environment is unconstrained, theimage that captured by the camera may affect bythe light of the environment. It will let the image provide inaccurateinformation and lead the process to misclassification. So, classifying based onmultiple images is much more reliable. Since the thesis focusing on the public area, if multiple audiences arestanding in front of the digital signage,the face of some audience may be messing up or cover by other audience whostand in front of him.As mentionearlier, this thesis focusing on pubic audience demographic, the faces thatcapture by the camera may possible to be small due to the distance between the camera and the audience.
The low-resolution image will get the smaller imageinput. So, the smaller the input of the image, the shorter time of theprocessing is required. In order to minimize the delay of the advertisement indigital signage, a good conduct will keepthe faces resolution as low as possible and the classification of the imagewill keep process for the next advertisement during the audience viewing thedigital signage. 1.
3 Research ObjectivesThe main objective of the thesis is to create aprototype to simulate the performance of a multimodal identification for agroup of audience in the camera view. In order to achieve it, the followingobjectives are outlined:o To design a new framework fortargeted advertising in public displayo To develop an adaptiveappearance model for monitoring the audience demographico To define a representation setof measurement for investigating the performance of targeted advertising. 1.4 ScopeReal-time audience demographic application wasdeveloped for digital signage that combinesthe LCD screen and digital camera. The digital camera will capture the video of the audience who stand infront of the digital signage. By using the computer vision method, theapplication performs face detection andtracking method to detect audience face image who stand inside region ofinterest and viewing on the screen.
After the audience face was detected, theapplication will perform several classificationsfor gender, age, glasses and beard detection. The database will include thetime of the advertisement, gender, age, beard and whether the audience wearingthe glasses or no. The result will be saved in the database as audience demographic. Besides that, the applicationalso will perform decision about broadcasting what category of an advertisement according to audiencedemographics. 2.1 Digital SignageDigital signage is nowgaining popularity and ubiquitous on advertising. It can display image,message, animation, video, and text.Digital signage consists display devices as liquid crystal display and displaycontroller as PC or digital media player.
It can be found in public and high populationareas such as shopping mall, retail shop,restaurant, bus station and corporate building. Digital signage has become animportant platform for deliveryinformation, message, and announcement tothe public. Figure 2.1 shows the simpledeployment of digital signage.Most of the organizationhas changed from conventional signage todigital signage. Conventional signage is a traditional advertising method that delivers information at a specific location.
Conventional signage hasmany disadvantages on time-consuming andcost of the construction and installation of the signage. The otherdisadvantages of conventional signage areit only can display static information such as image and text. Digital signagereduces the cost of construction and installation of the conventional signage.It able to display image, animation, video, text and interactive interfaces.
Moreover,it also supports live video and webcontent. With the evolution of thetechnology, some digital signage had come with touch screen, body sensor or QRcode via smartphone. It allows theaudience to interact with the digital signage system.
It will increaseaudience’s user experience of interacting with the advertisement of the digitalsignage (Using Interactive Digital Signage to Increase Customer Engagement, 2017). Networked digital signage is controlled via a computer networks such as internet or intranetand connects to the digital display. Figure 2.2 shows the networked digital signage. 2.2 Targeted AdvertisingTargeted advertising is an advertisement that used to serve a specificaudience which can be a group of people, individual or a particularaudience demographic.
Instead of promotingadvertisement to the public, targetedadvertising allows the advertisement to reach to the right people asmany as possible. Digital signage also appliesthe targeted advertising to increase the effectiveness of the digital signagecontent (Why is digital signage effective?, n.d.). This enables the digital signage content able todeliver a message to the right audienceat any time. 2.2.
1 Targeted Advertising in OnlineAdvertisingOnline advertising also knows as online marketing. It is a marketingplan and advertising which allow the user to promote their product or serviceto consumers through the internet.Facebook and Google were the examples ofthe online advertising that was usingtargeted advertising. With using targeted advertising, Facebook able to let theadvertiser promote their product or service to the targeted consumer based onthe location, age, gender or keyword that consumer search on Facebook.There are three type oftargeting advertising methods:· Behavioural Targeting – Theadvertiser uses the activity and action of the user such as the pages that theyvisit or search to determine which advertisement to show.· Content and ContextualTargeting – The advertiser targets people based onthe content people are reading on the websitefor keyword and show advertisements to the webpage based on the keyword.
· Geographic targeting – Theadvertiser target user based on the geographic location. Internet protocoladdress (IP address) can be signal the location of the user. After locating the location of the user, it willdisplay the advertisement based on their geographic location.
2.2.2 Targeted Advertising in DigitalSignageTo establish targeted advertising on digital signage, the human face is required for face detection andface recognition. During the advertisement played, the camera was used to capture the audience’s face image who look atthe digital signage. The system will detect the audience’s face using Haarcascade and pass the face image into the frameworkto extract the demographic data. After that, the result of the data will be used to find out a right advertisement to play to the audience.Figure 2.
3 shows the flow of the targetedadvertising in digital signage. Figure 2.3: Thebasic flow of targeted advertising in digitalsignage The human face is a source of information about the attributes of aperson such as age, gender beard, skin color,and glasses. Face detection can be used to identify the face multimodalattributes and develop an adaptiveappearance model for monitoring the audience demographic.
2 It enables digital signage to increase theeffectiveness of targeted advertisement played inpublic area. 2.3 Face DetectionThe process of extracting the human faceimage from image or video from the real-time application. Face detection findsout the location and the size of the face image by using Haar Cascade feature.Haar cascade feature is an object detection method that proposed by Paul Violaand Michael Jones in 2001. 1 Nowadays, this feature method still used widelyin face detection research.
It is a machine learning method that trains theimage inside the database and used to find out and locate the object in anotherimage. The images that used for trained will be separated into two types:positive and negative images. For face detection, positive images are the imagethat includes the human face and negative image is the image that not includeany human face image. After training the positive and negative images, themachine learning will update the classifier and the classifier are able to findout and detect the face image in another image by extracting the feature. If theperformance is weak, that is the high false positive and false negative rates.2.3.
1Face Tracking After the human face was detected, we wantto track the face of the person with using Continuously Adaptive Mean Shift (CamShift)method. 4 CamShift is the method that uses for object tracking and tracks thecolor object with the combines of the mean shift. There is the problem of theCamShift is when there has the same color object that covers the color imagethat was tracking by the Camshaft. Face area can change according to the colorof the image that track.The step of the Camshift algorithm: 41. Set theregion of interest (ROI) of the image. 2.
Selectthe initial location of the search window with Mean Shift. Example: face imagethat detects by Haar Cascade3. Calculatethe color probability of the center of the ROI.
4. Using meanshift to find out the centroid of the image and store it as the 0th movement.5. For thenext frame, repeat step 3 and step 4 until the mean shift location moving thresholdwas less than a preset threshold.2.3.
2Local Binary PatternLocalBinary Pattern (LBP) is an approach method used for texture and image analysis.Local Binary Pattern was first described by T.Ojala in 1994. 3 It has beenuse as the powerful feature for textureclassification.
Until now, it not only a method that used for textureclassification and it also becomes a method that used for face detection andrecognition.