Various government departments and commercial organisations require information in regard to motor vehicle activity; this includes statistics on traffic movements as well as information on particular vehicles. This information is used to plan future road works, monitor car parks, adjust the timing of traffic lights or even to identify and determine the speed and behaviour of individual vehicles. The data requirements vary according to the particular needs of the information system, as do the technologies used by these systems. Let us examine examples of systems that collect data using air pressure, magnetic fields and video.
Systems based on air pressure sensors are used to count the number of vehicles passing a given point. These systems include three major components; hollow rubber tubing, air pressure switches and a control unit, each of these components are visible in Fig 3.25. The hollow tube is placed across the roadway and is connected to an air pressure switch on the control unit; the other end of the tube is sealed.
As a vehicle’s tyre runs over the tube, the air pressure within the tube increases; this increase in air pressure activates the air pressure switch.
The controller detects the change of state of the air pressure switch and increments a counter.
Obviously a car will activate the pressure switch twice, and large trucks more than twice. Most controllers merely divide the number of activations by two, this is sufficiently accurate for the majority of applications. Many controllers contain multiple air pressure switches; this allows more than one traffic lane to be monitored using a single controller. Commonly the controller contains a timer, so each activation actually results in a time being recorded by the controller. The data held in the controller is transferred to a computer for further processing.
Other systems use induction coils to detect changes in magnetic fields; commonly these systems are used to control traffic lights at intersections. A loop of wire (the induction coil) is installed permanently within the surface of the road (see Fig 3.26). As a vehicle enters the loop, the vehicle’s magnetic field alters the current flowing through the wire loop; this change is detected by the controller. These systems, rather than counting vehicles, are generally used to determine if a vehicle has arrived at a red traffic light. In practice this means the lights can remain green in the major traffic directions and only change when a vehicle is detected on the minor road.
Fig 3.25
Air pressure based vehicle counter, with inset air pressure switch.
Fig 3.26
The dark lines on the above roadway indicate an induction loop is installed.
GROUP TASK Discussion
Both of the collection devices above have limitations. Discuss scenarios where each device may fail to correctly identify a vehicle.
Systems utilising video are able to do far more than just count the number of vehicles;
they can also identify individual vehicles, monitor driving behaviour and calculate speed. Examples of such systems currently in operation in Australia include speed cameras, bus lane cameras and heavy vehicle monitoring cameras.
Let us consider the Safe-T-Cam system, developed by the CSIRO, and used in NSW to monitor heavy vehicle movements. The aim of this system is to record the registration number and time as each heavy vehicle passes each Safe-T-Cam. As there is a network of more than twenty Safe-T-Cams around the state this data can be used to track the movement of individual vehicles.
Each Safe-T-Cam uses a video camera and a still camera as collection devices. Frames from the video being used to track individual vehicles and detect when they are in the correct position for the still camera to take a photograph.
The system identifies an individual vehicle by comparing the differences between the current video frame and an image of the stationary background; these differences are used to produce a difference image (see Fig 3.27). An individual vehicle can be tracked by examining the difference images produced from a sequence of frames. The example frame shown in Fig 3.27 contains just one vehicle, obviously it is more likely that multiple vehicles will be contained within each frame, the system must be smart enough to identify the same vehicle as it appears in subsequent frames. This is accomplished by comparing various attributes of the vehicles on each difference frame; if the attributes are similar then presumably it is the same vehicle.
When each identified vehicle reaches a set point on one of the video frames the digital still camera takes a photograph. These photographs are of sufficiently high resolution to allow the registration number on the vehicle to be read using optical character recognition (OCR) software. Finally the registration number and time of detection is transmitted to a central monitoring site.
Current frame Stationary background
Difference frame Fig 3.27
Safe-T-Cam compares the current frame with a background image to
produce a difference image.
Fig 3.28
Sample photo, typical of those collected by Safe-T-Cam. The registration number can be clearly determined.
GROUP TASK Discussion
Traffic authorities aim to ensure heavy vehicles are registered and that drivers only drive for a specified number of hours each day. How can the data collected by Safe-T-Cam assist in achieving these aims? Discuss
GROUP TASK Research
There are various systems being developed that not only monitor vehicle activity but are actually able to control vehicles. Use the Internet to research these new tecnologies.
(a) Outline the technology and processes occurring as a digital camera collects and digitises image data.
(b) Outline the technology and processes occurring as a flatbed scanner collects an image.
(c) Outline the technology and processes occurring as sound is recorded using a microphone, sound card and computer to produce a sampled digital file.
Suggested Solution
(a) At the back of a digital camera is a CCD (or a CMOS chip) which contain a photocell for each pixel. The photocells respond to the amount of light falling on them through the lens of the camera. As the photocells in a CCD are hit by incoming light, they emit a current relative to the brightness of the light. Different colours are detected through the use of a red, green and blue bayer filter covering the photocells. The electrical current is converted using an analog to digital convertor into an equivalent bit pattern for each pixel. The digital data is then processed to generate a red, green and blue colour value based on the values of the adjoining pixels. Finally the digital image data is compressed and stored on the camera’s flash card.
(b) A light source and row of sensors are mounted on a head assembly that moves in precise steps over the entire image being scanned. The more light that is reflected, then the lighter that part of the image must be. A series of focusing mirrors and lenses focus the reflected light onto the row of light sensors, which is typically a CCD. As the reflected light hits the sensors a current is produced – the more light, the stronger the current. An ADC (analog to digital converter) then converts this electrical signal into an equivalent series of bits for that section of the image. For colour scanning, there are 3 coloured lights, one for each of red, green and blue light. Each coloured light flashes in sequence over each line of the image, hence the CCD collects the intensity of red, green and blue light. The equivalent bit patterns representing the colour of each pixel of the image can then be determined.
(c) Sound waves are detected by the microphone as a sequence of compressions and decompressions in the air. These movements in the air move a diaphragm inside the microphone which in turn generates an equivalent current. This analogue current is sent along the connecting cable through to the sound card. At the sound card, the current is sampled many thousands of times a second and put through a ADC. Each sample represents the height of the original sound and is converted by the ADC into a binary number. This effectively converts the sound wave into a long sequence of binary numbers. Most sound card also contains a digital signal processor (DSP) which smooths the transitions and filters out noise. The resulting sound samples are then transferred via the system bus to the CPU where they are directed to a storage device.
HSC style question:
SET 3B
1. The main difference between CCDs used in digital cameras and those used in flatbed scanners is:
(A) digital camera CCDs have a single row of photosites, flatbed scanners have a two dimensional grid of photosites.
(B) flatbed scanner CCDs have a single row of photosites, digital cameras have a two dimensional grid of photosites.
(C) flatbed scanners generate their own light, digital cameras use natural light.
(D) digital cameras generate their own light, flatbed scanners use natural light.
2. A bayer filter:
(A) is used to remove unwanted detail during image collection.
(B) converts analog data into digital data.
(C) alternates red and green rows with blue and green rows.
(D) is used to compress image data.
3. Microphones are used to:
(A) convert sound waves into bits.
(B) convert sound waves into electrical energy.
(C) convert electrical energy into bits.
(D) All of the above.
4. The component on a sound card that filters and compresses the audio data is known as a:
(A) ADC (B) CCD (C) SAR (D) DSP
5. CCDs that contain two layers of sensors are commonly used in:
(A) analog video cameras.
(B) digital video cameras.
(C) digital cameras.
(D) Both (A) and (B).
6. The main components of a dynamic microphone include:
(A) two plates and a power source.
(B) a diaphragm, wire coil and magnet.
(C) a capacitor, comparator, DAC and SAR.
(B) The data is collected to suit normal analog television.
(C) The entire contents of the CCD is read for each image collected.
(D) They use interlacing to reduce the amount of data.
8. Mechanisms on traditional cameras to control the amount of light entering the camera include:
(A) altering the size of the aperture and the time shutter is open.
(B) altering the size of the shutter and the time the aperture is open.
(C) the use of different types of flim that have varying sensitivities to light.
(D) moving the lens in and out to focus the light more accurately.
9. One issue to consider when using vehicle counters based on a single air pressure switch is:
(A) the road temperature and air temperature commonly cause false readings.
(B) they are unable to produce digital data.
(C) they cannot differentiate between vehicles with two axles and larger vehicles with more than two axles.
(D) the vehicle must be stationary if the magnetic field is too influence the induction loop.
10. The Safe-T-Cam system uses a video camera and a still camera. Why is this?
(A) Still images are used to isolate individual vehicles from the
background, video is used to determine the registration.
(B) Video is used to isolate individual vehicles from the background, still images are used to determine the registration.
(C) The still images are used as a backup should the video camera fail.
(D) This is not true, Safe-T-Cam only uses a video camera..
11. Describe the operation of a digital camera.
12. Describe the operation of a condenser microphone.
13. Explain how an ADC performs its conversion using the services of a DAC.
14. Explain the differences between CCDs used in digital still cameras and those used in video cameras.
15. Research and describe how speed cameras operate.
SOFTWARE USED FOR COLLECTION