Assignment no .5 Image Processing Intelligent Urban Traffic Control System (KKKA 6424) Supervisor Prof. Dr.

Riza Atiq Abdullah OK

Prepared by: SARAH HAZIM

P65407 RASHA SALAH P64799

Introduction
Crime rates are much higher in big cities than in either small cities or rural areas, and this situation has been relatively pervasive for several centuries. Higher pecuniary benefits for crime in large cities can explain approximately 27% of the effect for overall crime, though obviously much less of the urban- crime connection for non-pecuniary crimes. Lower arrest probabilities, and lower probability of recognition, are a feature of urban life, but these factors seem to explain at most 20% of the urban crime effect. The remaining 45-60% of the effect can be related to observable characteristics of individuals and cities. A third unintended consequence is the possibility that there will be an increase in recorded crime for some crime types with Increased Crime Reporting. Many offenses have low reporting rates, especially minor acts of violence, graffiti, and drug offenses. CCTV operators are better placed to spot these offenses and this can actually drive up their recorded crime figures, as happened with narcotics offenses in Oslo Central Train Station23. This is not to say there was an increase in actual crime, just recorded crime. This is a potential outcome, and you may need to prepare other people involved in a future CCTV system of this possibility.

How CCTV Aims to Prevent Crime

A CCTV system is not a physical barrier. It does not limit access to certain areas, make an object harder to steal or a person more difficult to assault and rob. This does not mean it is not an example of situational crime prevention. It is highly situational, and as will be shown, does have some crime prevention capacity in the right situations. Although CCTV has many functions, the primary preventative utility is to trigger a perceptual mechanism in a potential offender. It seeks to

change offender perception so the offender believes if he commits a crime, he will be caught. In other words, CCTV aims to increase the perceived risk of capture, a factor which, assuming the offender is behaving in a rational (or limited rational) manner, will de-motivate the potential offender.

Video Motion Detection

Introduction There are many methods of detecting intruders into premises. These include such systems as:

Intruder alarms. Fence mounted detectors. Buried vibration or electric field devices. Active infrared devices. Passive infrared devices. Microwave devices. Video motion detection devices.

Video motion detection device (VMD)

Video motion detection (VMD) is a way of defining activity in a scene by analyzing image data and differences in a series of images. The functionality can either be built-in into a network video product or made available with video management software. In-picture alarm programming allows you to define areas of a screen where you want to detect any visual changes.

VMD in DVR (digital video recorder) systems

in such a system, cameras are connected to a DVR, which performs the VMD on each video stream. This allows the DVR to decrease the amount of recorded video, to prioritize recordings and to use motion in a specific area of the image as a search term when searching for events. The downside of this method is that performing VMD is a CPU intensive process and performing VMD on many channels puts a heavy strain on the DVR system.

VMD in network video system

VMD, as an integrated function of network cameras or video servers, offers substantial advantages over the scenario mentioned above the most significant

being that the VMD is processed in the network camera or video server itself. See event handling in a video server.

A Network Camera With Built-In Vmd Feature

This alleviates the workload for any recording devices in the system and makes "event-driven surveillance" possible; that is, unless activity is detected in the scene, no video (or only video with low frame rate) is sent to the operator or recording system. If motion is detected, you may request a number of things to be done automatically; e.g. save images before, during and after the event (called pre- and post-alarm image buffering), start sending images for recording at specified destinations, send alerts to specific individuals via e-mail or SMS to a mobile telephone, activate lights, turn alarm on or off, open or lock doors, control temperature levels etc. VMD data with information about the activity can also be included in the video stream to simplify activity searches in recorded material. VMD can also reside in the video management software, thus providing VMD functionality to network cameras that do not originally embed this feature.

Sample of the motion detection user interface offered in AXIS Camera Station software.

This report is concerned with Video Motion Detection devices. (VMD). these may be within or outside the premises and, besides detecting intruders, can be used as part of a building management system. VMD may often be used either as a standalone system or integrated with other detection systems. In an ideal world, detection devices would give no false alarms and 100% of genuine alarms. Unfortunately, this is not an ideal world, and a certain amount of compromise is necessary. This compromise must be reduced to the most effective and acceptable level to achieve the system objectives. Video Motion Detection is an electronic method of detecting a change in the field of view of a camera. In its simplest form, this is achieved by storing one frame of the video information and then comparing the next frame with this to decide whether there has been a change. The change detected would be a difference in the video voltage, indicating a change of brightness within the scene. This would be initially ignored as an alarm until a further frame confirmed the change, or not. If confirmed as a change of brightness in the scene, then an alarm would be generated. This could cause a contact to close and activate some warning device such as a buzzer, or cause the switcher to select the camera that detected the motion. The sampling process may take somewhere between one fiftieth of a second and one second to detect a change, depending on the method of sampling. This simple detector could be used in an environment where all conditions were absolutely stable and the only possible change in brightness would be due to an intruder. However, the intruder could be a mouse or a person. The system couldnt differentiate between the two.

Video sequence consist of still picture images which are visualized and interlaced one after another to make it as a video [1, 2]. The images changes convey different information in two dimensional are produce rapidly which unable the humans naked visualization system to detect or track the motion translation in the moving images [3]. Due to this reason, there is a necessity to analyze the image sequences which are taken using a camera. To analyze the motion translation, Block Matching Algorithm (BMA) is the simplest technique used where a frame is divided into small block size 16 16 [4, 5] or 8 8 pixels blocks [6]. Each blocks from both, current image and previous image is compared based on the pixels value to analyze the motion translation. This paper presents a simple block based pixel subtraction technique to analyze the motion translation between both images based on area of interest. The motion translation analysis is conducted based on 16 16 block size. Block of area of interest from each image is selected and pixels value of both selected blocks is subtracted to analyze the motion translation between two selected blocks. Non-zero pixels value represents

the motion translation and zero pixels value represents no motion translation in the analyzed blocks.

Benefits of CCTV
Crime is still prevalent in a country as safe as Singapore, thus we must never give up on the fight to deter or prevent crimes. This can be done easily with the implementation of CCTV at housing estates, shops and offices. Here are the reasons why crime can easily be deterred because of a CCTV system:A CCTV system can prevent a crime by deterring potential criminals with the physical presence of a camera. This can be done with presence of cameras and CCTV monitors to staff at cashiers and the front of house staff. This prevents crime committed by criminals who do it on impulse and greatly reduce the numbers of shoplifters and car theft as fear strikes them at the core and thus paralyzing them from committing the crime. Not only does it deter unplanned crime, but it also gives a sense of security and safety to your other customers and employees knowing that everything that happened at the location is recorded.The next thing a CCTV system does is to filter out potential criminals. In the case of shoplifting, a well-trained staff may see the anxiety of a potential shoplifter and deter the potential criminal from committing the crime by approaching the potential criminal and discouraging him personally before he even commits the crime. This can also be applied to crimes committed in highly secluded areas like car-parks and the security personnel may accompany overtime staff at office building to their cars if he notices any suspicious person or activity at the car-park. Seeing that someone else is in the vicinity, crime involving rape, murder and robbery which happens when one is alone can thus be avoided.

Even if the crime occurs future perpetrators will avoid committing the crime at a

place where it seems that all the criminals paid highly for their crimes. This could be done in conjunction with staff putting up reports and photographs depicting the criminals paying for their crime. Likewise, this is done to deter opportunist criminals and possibly reduce the crime in that area. The news could also leverage on this by releasing footages of a crime which happen which will not only lead to a faster arrest on the criminal, but also warn other like-minded people that their wrongdoings are caught on footage. The public thus becomes more educated in spotting crimes and they will also become informants for the relevant authorities as well.When a CCTV system is set up with a validated security firm, these two will form a system that will break down the open opportunity of crime even more. With cameras around remote areas acting as watching eyes of areas not covered, a trained security guard may inform the guard on patrol of any suspicious activity in another area. As such, the CCTV plays the role of an informant to the security firm who will take the appropriate action to stop or prevent a crime from happening.

In order for a CCTV system to work to its maximum advantage, it is important to note 2 things; location and people. Location is extremely important to the placement of the cameras as well as the surveillance area. Trained personnel can react better to the situation on the ground if the surveillance area is close by to all the areas being monitored. That way, potential crimes can be prevented swiftly and also the criminals arrested swiftly when deterrence fails. People are extremely vital in preventing crime as more eyes on the ground means an earlier detection of crime. This could mean a potential crime is avoided all together which not only ensure the place is safe, but also a second chance to criminals who have plans for those locations to abort their plans and seek help if necessary. The offender must be aware of the cameras' presence.

An important consideration in the effectiveness of a surveillance technology is the type of crime to be tackled, because this impacts the criminals' ability to adapt. Although a CCTV system may reduce the likelihood of burglary at a commercial location within the range of the camera, there is some evidence that drug markets can continue operation in the presence of CCTV by changing their operating practices. For example, at one location some offenders met and discussed business in the cameras' presence, but concluded the transaction at another site12. In other CCTV areas, however, drug crime that could not successfully relocate or adapt to the cameras was eradicated. Fake cameras have been employed in some instances. Poyner13 reports that crime was reduced on public buses after the installation of both active and dummy cameras onboard a number of buses (indeed crime reduced on more buses than the ones fitted with any cameras, a concept known as a diffusion of benefits). It is therefore possible that fake cameras could achieve the same preventative aim as active systems. However, if users of the space under surveillance are led to believe - through signs, for example - that they are being watched 24 hours a day and an incident occurs, the misrepresentation of a form of guardianship may have liability implications. A third, more general mechanism by which CCTV may reduce crime is through an increase in collective efficacy. Welsh and Farrington14 argue that if residents see CCTV cameras being installed in their neighborhood, this will signal to them a degree of investment in and efforts to improve their local area. They argue that this might lead to greater civic pride and optimism, and, as a result, lead to an increased level of informal social control among the local people. A counter to this argument is that overt cameras may instead lead to a neighborhood being labeled as highcrime, accelerating the process of social disorganization. Other Benefits

A number of other benefits, beyond a reduction in crime, may be accrued from a CCTV system, including: Reduced fear of crime Aid to police investigations Provision of medical assistance Place management Information gathering Diffusion of benefits The following section describes these potential benefits in more detail.

Reduced Fear of Crime Numerous studies have tried to determine if the presence of cameras in public places reduces fear of crime in people who use the area. These studies, many of which interviewed people in the CCTV area, have examined whether consumer buying has increased in areas with new CCTV systems. The general argument is that the area will benefit from a positive economic impact when people feel safer. The findings are mixed but generally show there is some reduced level of fear of crime among people in CCTV areas, but only among people who were aware they were in an area under surveillance. Most studies exploring the perception of surveillance areas found that less than half the interviewees were aware they were in a CCTV area. Reduced fear of crime in an area may increase the number of people using the area, hence increasing natural surveillance. It may also encourage people to be more security conscious.

Aid to Police Investigations

Diffusion of Benefits
Although rarely addressed in the research literature, there is also the distinct possibility that if offenders are aware and cautious in the presence of cameras, they may be unaware of the extent of the cameras' capabilities. As a result they may curtail their criminal activity in a wider area than that covered by the camera system. In effect, this extends the value of the cameras beyond their area of operation, a process criminologists call a diffusion of benefits.

Pixel analysis
Motion translation is rapidly evolving field for several applications which include remote surveillance, remote working and learning and etc. It is also can be defined as an upcoming information and technology based research in image processing area. Motion translation also helps into conducting motion analysis in terms of providing information in video processing and computer vision applications. In competitive schemes, many motion estimation algorithms are developed to analyze the images from video sequences. In this paper, the proposed idea is to determine the motion translation between two selected blocks of current image and previous image in a video sequences, the video is extracted into single image as illustrated in Fig. 1. The video is extracted from the first image (N image) to the last image (N+1 image). 60 2011 International Conference on Signal, Image Processing and ApplicationsWith workshop of ICEEA 2011IPCSIT vol.21 (2011) (2011) IACSIT Press, Singapore Two targeted images (current image and previous image) are selected from a video sequences which consists of many images. The current

image and previous image is then divided into small blocks size of 16 16 as shown in Fig. 2. Both selected images are analyzed based on the pixels values.

The small blocks at the current image and previous image is compared with each other as illustrated in Fig. 3. Equation (1) is applied into current image and Equation (2) is applied into previous image is used to compare each block division. Each divided block is used to subtraction between the current image and previous image. Each blocks pixels are subtracted to analyze the pixels values by using Equation (3).

Fig. 4 explains the proposed experiment in flowchart. A surveillance video is possessed from the Security Department of Universiti Teknikal Malaysia Melaka to test run the developed video coding. The 61flowchart explains the full experiment setup step by step. The flowchart can also be the reference to develop the pixel subtraction technique for future extension.

3. Experiment Result and Data Certain condition need to be persistent in order to analyze the changes of pixels value in the selected images. This is to ensure that the same environment and condition is used for each image during the conduct of motion translation. The condition which is set constant is as follows:

 Block size of 16 16 pixels Surveillance video (frame size of 128 160 pixels In this experiment, a surveillance video is applied to conduct the motion translation analysis in an image. Fig. 5, Fig. 6 and Fig. 7 shows the extracted data of three different types of images in the experiment. In the surveillance video sequences in Fig. 5, three different images are selected for data analysis. In the first analysis, previous image (a1) and current image (b1) shows no motion translation information. The selected coordinate block at (4, 3) area of interest shows no motion translation as depicted in c1. The subtracted result is presented in matrix form as shown in (d1).

Fig. 7: Motion Translation of Lady Guard Head

In the second sequence, a static background is chosen for previous image (a2) while the current image (b2) shows a car passing through the surveillance video camera. A block is selected at coordinate (7, 4) as an area of interest to analyze. Both selected blocks are subtracted and the result is presented in matrix form as shown in (d2). Presented result in (d2) matrix shows a drastic motion changes in the pixels value at the left corner. This pixels value change represents the front portion of the moving car. In the third sequence, previous image (a3) shows a static background while the current image (b3) shows lady guard walking passes the surveillance video camera. A block is selected at coordinate (8, 5) at both images are subtracted. The result is presented in matrix form in (d3). The matrix (d3) shows that there is a high increment in the pixels value which is located at the bottom of the matrix. The changes occur represent the head of the lady guard.

Discussion
When the value of pixels is subtracted and produces zero values, no motion translation is detected. Thus, the images are carrying the same information as resultant in Fig. 5. Some of the pixels in the first sequence are not zero value due to the light intensity occurrence due to the sunny day. This reading can be ignore sue to the sun light intensity and there is no drastic impact towards the motion translation in the image. In Fig. 6 and Fig. 7, the pixels change represents the front portion of car and head of lady guard respectively. There is difference in the pixels value when both images are subtracted. This happens because different value of pixels represents different information and color. The matrix result (d2 and d3) presents areas with no object detected (such as car and lady guard), the subtracted pixels value are relatively zeros or at minimal value (due to the intensity of light). The rest of the area represented with pixels value is subtracted and motion translation occurs in that area.

Conclusion
Block based subtraction technique is a simple method in detecting, estimating and analyzing the motion translation that happens in a sequence of images. This technique reduces the processing time and memory saving because its only processes the area of interest.

Area of interest CCTV Monitored System Benefits:

CCTV was used in security systems and such applications as retail shops, banks, government establishments, hotel etc. The true scope for applications is almost unlimited. Some examples are listed below.

1. Deters theft, fraud and vandalism

2. Discourages workplace transgressions

3. Deters false accident claims

4. Protects employees and those on the premises

5. Monitoring traffic on a road.

6. Reduces the risk of theft or damage

7. The well-publicized use at football stadiums.

8. Reduces the need for on-site guards

References
[1] Khan, M.A., An Automated Algorithm for Approximation of Temporal Video Data Using Linear Bezier Fitting. The International Journal of Multimedia & Its Applications (IJMA), 2010. vol.2, no.2. [2] Martin Stanek, L.S., Unpuzzling Puzzle (Analysis of A Video Encryption Algorithm). 20th International Conference on Advanced Information Networking and Applications, 2006: p. 20-24. [3] Pekka Sangi, Janne Heikkila and Olli Silven, Extracting Motion Component From Image Sequences Using Particle Filters, The 12th Scandinavian on Image Analysis, Bergen, Norway, 2001 [4] Dhahri S., Zitouni A., Chaouch H., and Tourki R., Adaptive Motion Estimator Based on Variable Block Size Scheme, World Academy of Science, Engineering and Technology, 2009: pp. 384-390.