License Plate Recognition, What it is and How it Works

Products using License Plate Recognition (LPR) Technology are gaining in popularity all over the world. One draw is its convenience. Since all countries and states require registered vehicles to display at least one license plate, there is no need for a transmitter or transponder to be attached to the car.

 Police departments make use of the technology for traffic law enforcement............

and for auto theft recovery.

 Government entities use it for security, restricting vehicle entry to secure areas. State highway authorities, as well as private parking facilities, are it using to extract payment for toll and parking fees. How do you think Repo men track down cars whose drivers neglect their auto loan payments? 

Yes, LPR systems. 

Since we’re all under its scrutiny, it would be nice to know how it works.

LPR is a form of Optical Character Recognition. OCR is the technology that is concerned with the electronic recognition and digital encoding of printed or handwritten characters by means of an optical scanner and specialized software. In OCR processing the entire image is analyzed for light and dark area so each individual character can be identified as such and saved as a coded binary character instead of a picture.

The components of LPR are designed to provide optical character recognition (OCR). The results depend on two factors:

·         image accessing components -  camera and illumination

·         Image recognizing components –   software and the recognition algorithms built into it.

Let’s look at a typical system

Camera

 An  IP camera will deliver a quality image. This is especially critical in the case of cameras that capture fast-moving vehicles. The best systems combine a camera lens with a short shutter speed and high power illumination to minimize any motion blur which would compromise recognition accuracy. Depending on the objectives, a camera is positioned to capture either the front or the rear license plate.

In some applications more than one camera is used, such as when the driver’s face also needs to be captured

Illumination

Infra-red flash has been found to be the best type of lighting for LPR. First of all since the human eye cannot detect it, the driver will not be startled, nor will any covert applications be compromised. Second, most states are retro-reflective license plates which reflect I-R light quite well.

 And third, it works equally well in day and night light.

Frame Grabber

This is the interface which provides connectivity between the camera and the PC running the LPR software, allowing it to read the image information. The frame grabber converts the analog signal from the camera into a digital stream of 1’s and 0’s that can be understood by the computer.

Computer

This is usually a PC running Windows or Linux. The computer is the coordinator of the system – running the LPR application, reading the images, analyzing and identifying the plate, all while interfacing with other applications and systems

Software

License plate recognition technology falls into the category, OCR or Optical Character Recognition. OCR technology uses software that reads the text of scanned images. Computers use OCR software to convert scanned text into binary coded digital documents.LPR software, in particular, reads the text from digital pictures of a license plate and translates it into a form that the computer can manipulate.

Putting it all together

Let’s take a look at how a LPR system works when it’s used for access control.

·         A vehicle is approaching a lift gate. It activates the LPR system by driving over a magnetic loop.

·         Infra-red illumination lights up the targeted area, usually the front or back of the vehicle including the license plate and the camera takes the picture.

·        The Frame Grabber allows the image to be sent to the PC

At this point, let’s take a detailed look. The targeted area must be singled out, i.e., the license plate. Localizing is an algorithmic function that determines the portion of the image that holds the license plate. The first task of localization is to disregard the headlight, mirror, grill, bumper, etc.

What the algorithm is searching for is a rectangular shape. But since a vehicle can have several rectangles, additional algorithms must be employed to seek out characteristics endemic to the plate, such as background color and particular dimensions. (Since license plate size and dimensions are not uniform worldwide, some systems are country-specific.)

·         Once the license plate has been isolated, the software takes over, using a different set of algorithms. It enhances the image, locates the license plate, extracts the string of characters, and identifies the fonts.

 Again, a closer look is in order. Now that the plate has been localized, other algorithms come into play.

An algorithm to adjust any angular skew so the plate image can be accurately sampled, corrected and recalculated to the optimal size

Another algorithm to regulate the brightness and contrast.

An algorithm to separate the alphanumeric characters so each one can be processed by OCR algorithms.

The isolated string of characters is translated by an OCR algorithm into an alpha numeric text entry.

And lastly a complex set of algorithms to verify the alpha numeric information and sequence, dictated by a specific rule set.

 These algorithms operate within millisections with each one dependent on the completion of the one preceding it.

Database

Upon completion of the sequence outlined above, the resulting entries can be stored on a local database or sent over a network. Minimum information would include license plate numbers, vehicle make, time, date, location. In some instances the image of the driver’s face might be desired.

Personal Testimony

Last summer, I had the opportunity to be the subject of LPR. I was driving a rental car in Florida en route to Key West when my directions instructed me to get on what’s known as Florida’s Turnpike. Subsequent signs informed me tolls were solely electronic which translated into no cash tolls. Cars needed to display the Sun Pass.

I was familiar with "Passes". My own car parked back at JFK airport in NY had an EZ Pass transponder anchored on the windshield.

EZ Pass is an interstate system that allows for prepayment via credit card. With EZ Pass one can travel the Eastern Seaboard  from Maine as far south as Virginia without stopping to pay tolls. Every time a car passes a toll reader, the fee is deducted from the credit card payment.

 Once it is exhausted, the card is hit and your account is flush once more.

 Problem was – Florida does not honor EZ Pass. It only has eyes for the Sun Pass. My rental did not have a Sun Pass.

 I was taken back. I had asked the rental agent in Orlando if I’d run into this problem and he’d assured me all Freeways accepted cash. Now what? An instant answer loomed ahead. The second option was that cameras would record your license plate and you would be billed for your total tolls plus an administrative fee.

 I wondered how this worked. I assumed Florida would bill my rental company. Would the company then bill me? Or since they already had my credit card information, would they just go ahead and charge the card? To make matters more complicated, I was driving a car with Georgia plates. I wondered how long it would take for the process to play out.

Not too long. In August I noticed the rental company had placed a charge for $13. What a seamless operation. And now I know exactly how it occurred. LPR to my rescue.

Whether you're seeking  information about LPR products or IP cameras, contacting Kintronics is also a seamless operation whether you choose to go to our website www.kintronics.com or call us at 800-431-1658. One of our knowledgeable sales engineers will be glad to assess  your needs and recommend a system. Or better yet, just fill out this information form for a quick response.

 

The Lumiere Brothers - Bringing Moving Pictures to the Public

Welcome back to our blog series exploring the family tree of the IP camera. Our first entry focused on  Eadweard Muybrhttp://kintronicsvirginia.blogspot.com/2012/10/before-video-camera-first-steps.htmlidge who  based his zoopraxisope on the knowledge that individual still frames, when delivered at more than 14 frames per second, trick the mind into thinking it is seeing motion.

 Thomas Edison came along and saw inefficiencies in the zoopraxiscope and thought he could improve on it. His kinetoscope, the subject of our second series, used multiple cameras in capturing the images rather than the zoopraxiscope’s one, and it also used actual film as media.

Now we will look at a branch that  sprouted ………in France…….…. with the Lumiere Brothers, Auguste and Louis.

 Ironically, or shall we say, appropriately, Lumiere means light.

 Their father, Antoine, was a portrait artist who,sensing  a promising future for photography, started a business manufacturing photographic equipment. Both brothers had technological aptitude, especially Louis. While still a student, he developed the dry plate process, a techniqueused in photograph developing. Upon graduation from college, they both joined their father in the business; Auguste as a manager, and Louis, using the knowledge gained in pursuing his degree in physics.

The company had been devoting itself to still photography equipment, but all that changed in 1894 when Antoine was invited to a demonstration of Edison’s Kinetoscope in Lyons. He came back impressed with the technology but intent on improving it.

First, there was the size of the device. At four feet high and over two feet wide, its bulk forced The Edison Company to confine most of its exhibitions to the Black Maria Studio at Edison Labs in New Jersey. 

 And the peephole mechanism - with one person viewing at a time - limited the size of the audience.

After listening to their father’s ideas, Louis immediately set to work on a smaller device, one that would bring simultaneous viewing to a gathered audience.He came up with the Cinematographe. Weighing only 11 lbs, it was no bigger than the handheld cameras of the era so it was easy to transport and set up. Not only that, but it was a camera, printer, and projector all in one.

Louis had also found fault with the Edison's design of  pulleys and sprockets to keep the loop of film in  continuous movement. Instead he devised  a system of intermittent movement which he described in the preamble to his patent application as:

The basic property of this appliance’s mechanism is to act intermittently on a regularly perforated strip of film to transmit successive displacements separated by stationary periods during which photographic images are either exposed or viewed.

To grasp this more easily think of the sewing machine which also relies on intermittent movement. The feed mechanism advances the material, then pauses it so that stitching can be applied, before further advancing, more stitching and so on.

Lumiere was also inspired by fellow Frenchman Emile Reynaud and his theatre Optique which projected successive frames on a screen. He incorporated it for the first Lumiere screening, held at an industrial meeting in Paris on March 22, 1895. The film, titled Les Sortie des Oevriers des L’usine Lumiere, showed workers leaving the Lumiere factory.

 Satisfied with the favorable reaction to his  premiere, Lumiere applied for an English patent on April 18, 1895. With intellectual ownership rights assured, the  Lumiere Brothers held more private screenings, and then amidst a buzz of excitement, had their first public screening in April in Paris in December of 1896

The program consisted of ten short films whose total viewing time was  not quite 20 minutes.

Seeing that people were eager to experience moving pictures for themselves, the Lumieres opened four theaters of their own, one each in Paris, London, Brussels, and New York City.

Legend has it that one of their films, showing the emergence of a train from a tunnel, sent the audience screaming out of the auditorium, convinced that the locomotive would be bursting through the screen.

 By 1896, their catalog of titles numbered 358 and would grow to 1,000 in 1898 and 2,113 in 1903. At the 1900 Paris Exhibition, they held an outdoor showing on a 99 x  79 foot screen.Yet despite their growing fame, the Lumiere Company retired from the moving picture business and returned to  manufacturing.

Why the departure from the mushrooming craze of" the movies”? It was Louis belief  that people would become bored with things they could encounter in life just by going outside for a walk. He is quoted as saying “  The cinema is an invention without a future,”

Instead the Lumiere Brothers returned their efforts  to still photography work and in 1907, designed  the Autochrome plate which was a giant step in the direction of  practical processing of color photographs.

 Louis Lumiere may have been a genius when it came to technology, but he had a long way to go as a cultural visionary.

Today he'd be amazed to see the role moving frames would play in developing video technology for the purposes of surveillance and safety. And even though  IP cameras make no use of film, they do owe  the Lumiere Brothers a word of thanks.

If you want information about IP cameras call Kintronics at 800-431-1658 or visit us at Kintronics.com. For specifics, fill out a request info form and one of our engineers will get back to you.

The Movie Camera, Framing Motion

(Part 2 of a series)

Just how does one get a still picture to move? And who figured out how to do it?

 If I could go back in time and ask these questions, we’d hear an international chorus of me, me, moi, moi, as Englishman Eadweard Muybridge, American Thomas Edison, and Frenchmen, Louis LePrince and Louis Lumiere all raise their hands frantically.  And they all own a piece of the answer, for no single inventor could have done it without looking at the ideas of the others.

In the next two posts we’ll confine our discussion to Muybridge and Edison for it is their inventions that are in direct line with what we call the video camera today. And it is also amusing to note that judging from the lives they lived, neither one of these pioneers of moving images was able to sit still for long.

Before we start though, let’s keep in mind that a movie camera, (and an IP camera) is nothing more than a camera that captures many images in sequence and records them on media. What sets the images in motion? Your brain.

We are only able to process individual frames if presented to us at a maximum of 12 per second. Our brains hold these images for about 1/15 of a second.  So if frames are presented at 15 per second or faster, we will perceive them as continuous motion. Early movies, or moving pictures, as they were rightly called had a frame rate of 14 to 24 images per second.

With that basic concept in mind, our previous post, Before the Video Camera: First Steps, looked at Muybridge’s contributions, let’s see what Edison added.

Thomas Edison

Young Thomas Edison was home-schooled after twelve weeks in a one room schoolhouse with thirty eight other youngsters spread across grade levels. The sole teacher had little patience with the six year old’s unquenchable curiosity and endless string of questions. 

Educators of today might label him as having ADHD (Attention Deficit, Hyper Active Disorder). And indeed his checkerboard of early jobs does point in that direction. Peripatetic as he was, though, he never left a field without improving on its tools.

Prior to becoming involved in the moving picture quest, Edison had already left an impressive trail of inventions. 

•  The automatic repeater – a device for transmitting telegraph signals between unmanned stations which ultimately allowed telegraphers to translate code at their own speed and convenience.

•  The quadriplex transmitter which allowed the transmission of telegraph signals over multiple frequencies

• The carbon transmitter which improved   audibility of Alexander Graham Bell’s telephone 
• An improved stock-ticker
• An electric vote recording machine

          His more well-known inventions include

                                                            the phonograph

AND

                                                             the incandescent light bulb

By 1887, Edison recognized the need to consolidate operations. He and his growing team of assistants moved into a building in West Orange, NJ, establishing what some call the first research and development lab.

In 1888, Edward Muybridge visited with his zoopraxiscope and a proposal for Edison. Why not  combine the zoopraxiscope and the phonograph? Initially, Edison liked the idea but the more he looked at the zoopraxiscope, the more inefficient he found it to be. He especially disliked that it depended on images obtained with multiple cameras, feeling it would be more practical and cost-effective if one camera could record successive images of a body in motion. Thus was born his next quest, or as some suggest, the next quest for his assistant, WKL Dickson. Doubts remain to this day how much Edison actually put into it, and how much Dickson did.

No matter who had the ideas, Edison was determined that Edison Labs take credit. Anxious to protect his invention-in-progress, he filed a caveat at the Patent Office in October, 1888, describing his idea for a device that would “do for the eyes what the phonograph does for the ear".

Edison called it the kinetoscope, taken from the Greek kineto for movement and scopos for watch. The initial kinetoscope did borrow from the phonograph, using a cylinder upon which tiny photographic images were affixed in sequence. When the cylinder war rotated, the illusion of motion was created, using reflected light. He subsequently discarded this version on the grounds of impracticality and began to look to the ideas of others engaged in the same quest.

French physiologist Étienne-Jules Marey had his chronophotographe which used a continuous roll of film to project still images. Edison found the film to be of insufficient length and durability.

 John  Corbutt was developing sheets of emulsion-coated film. When the Eastman Company produced their own version of the film, Edison bought up large quantities.

 In 1890, Dickson and another assistant, William Heise, came up with an improved kinetoscope that advanced a strip of film using a horizontal feed mechanism. On May 21, 1891 they demonstrated a prototype Kinetoscope that used 18mm wide film and functioned as both a camera and a viewer. When used as a camera, a rapidly moving shutter exposed the film at an intermittent rate; when used as a viewer the spectator looked through the same aperture and saw rapid intermittent views of the positive print. As explained at the outset of this post, when frames are presented at such a rapid rate, the mind sees them as being in motion. 

On August 24 of that year, the Edison Company filed a patent for a Kinetoscope using 35mm film. When completed in 1892, the Kinetoscope consisted of wooden cabinet, standing 4 feet tall, with a peep hole and magnified lens at the top.

 Inside, a fifty foot length of film with sprocket holes punched along each side, wound in a continuous loop around a series of spools, advanced by  an electrically-driven sprocket wheel at a steady rate.

 Beneath the film, a rotating shutter with a narrow slit opened and closed in rapid succession to admit a brief flash of light from an electric lamp under the film. So brief was the flash of light that the frame appeared to be frozen. However, the frames passed so quickly that the observer peeping through the lens saw a jerky but moving picture. 

A later version kinetoscope used a vertical feed, and it was this model that was used in the first public debut, a demonstration that took place on May 9, 1893 at The Brooklyn Institute of Arts and Sciences.

Two inventors, C. Francis Jenkins and Thomas Armat invented a film projector called the Vitascope and asked Edison to supply the films and manufacture the projector under his name. True to Edison’s penchant for putting his own stamp on his work, he declined and soon afterward, the Edison Company developed its own projector, calling it the Projectoscope. 

On April 23, 1896, in New York City the first motion pictures to be shown in a "movie theater" drew a capacity crowd. And with it, the door to inexpensive entertainment opened.  Theater was no longer reserved for the rich. 

Over a century later we can say the IP camera owes its basic concepts to Muybridge and Edison. For a moving tour of this evolution check out our video.

If you are interested in purchasing an IP camera or learning about IP camera systems, visit our website, www.kintronics.com

Before the Video Camera: First Steps

Just how does one get a still picture to move? And who figured out how to do it?

 If I could go back in time and ask these questions, we’d hear an international chorus of me, me, moi, moi, as Englishman Eadweard Muybridge, American Thomas Edison, and Frenchmen, Louis LePrince and Louis Lumiere all raise their hands frantically.  And they do all own a piece of the answer, for no single inventor could have done it without looking at the ideas of the others.

In this and the next  post  we’ll confine our discussion to Muybridge and Edison, for it is their inventions that are in direct line with what we call the video camera today. It is also amusing to note that judging from the lives they lived, neither one of these pioneers of moving images was able to sit still for long.

Before we start though, let’s keep in mind that a movie camera is nothing more than a camera that captures many images in sequence and records them on media. What sets the images in motion?

 Your brain; we are only able to process individual frames if presented  at a maximum of 12 per second. Our brains hold these images for about 1/15 of a second.  So if frames are presented at 15 per second or faster, we will perceive them as continuous motion. Early movies, or moving pictures, as they were rightly called had a frame rate of 14 to 24 images per second.

Keeping that basic concept in mind, let’s see what Muybridge did with it.

Eadweard Muybridge

British expatriate Eadweard Muybrige might never have been inspired to explore the moving image had not  Leland Stanford ,governor of California, businessman, and horse-owner been set on winning an argument.

Does a horse, at any time, while trotting, lift all four feet off the ground?

Stanford was on the yay side. In fact he was so entrenched that he hired Muybridge to prove him right. Famous for his large photographs of Yosemite Valley, Muybridge was a brilliant but eccentric photographer. And true to his nature, his new project followed a circuitous route to its completion.  Initially hired in 1872, Muybridge’s first attempts failed due to the lack of a fast enough shutter on his camera. His second attempt was delayed by six years; a period in which he was acquitted of murdering his wife’s lover.

After such a close brush with being confined to one space, he spent the next few years traveling though Mexico and South America. He supported himself with publicity photos taken for Union Pacific Railroad, owned by none other than Leland Stanford.

Upon his return to California Muybridge resumed his horse-in-action quest, working with a set-up of anywhere from 12 to 24 cameras, each equipped with a special shutter he designed to give an exposure of 2/1000 of a second. When lined up in sequence, it did appear that there were frames that captured  all four feet drawn up under the horse. Line drawings of his images soon circulated among the horsey set.

However with the publicity, came skepticism.  Doubters pointed to the leg positions in several of the frames and claimed they were anatomically impossible. Never one to back down from proving his point, Muybridge invented a device called a zoopraxiscope and took to the road for a series of lectures.

The zoopraxiscope was a lantern-like device that centered around a glass disc upon which he printed his photographs.  

 When the disc was rotated, the images were projected to the screen in rapid succession, giving viewers the illusion of motion. 

Some claim the zoopraxiscope was the precursor of modern cinema.  

Muybridge next devoted his efforts  to showing  humans in motion. His studies resulted in over 100,000 images capturing progressive movements within fractions of a second. 

A  question occurred to him : What if I could add sound?

 Upon hearing of the new sound producing phonograph Thomas Edison had invented, Muybridge and the zoopraxiscope traveled to New Jersey with just this proposition for Mr. Edison. 

Our next post will delve into what Mr Edison thought. Meanwhile if you are looking to purchase an IP camera or IP camera system or need information about one, please visit  www.kintronics.com