Computer Vision in the Retail Environment

In one research initiative underway, the Latin American and Caribbean Collaborative ICT Research Federation is looking into the use of computer vision for a real-time system that would allocate and supervise cash registers at grocery stores. Researchers believe that the service industry is one sector that could benefit immensely from new applications of computer vision. Their research project is aimed at improving service at grocery checkout lines by creating an automatic system that would monitor waiting times in lines, situations that create delays and required time necessary to serve the customer. This information and other measurements would be factored into the system so it can recommend the ideal number of check-out points. Continuous real-time measurements and the decision-making aspects of the system make it unique from the traditional passive measurements of existing computer vision applications.

Computer Vision in Farming and Agriculture

Another research project in progress is applying computer vision to automate agricultural machinery. Researchers at CLAAS Agrosystems in Germany are combining 2D and 3D computer vision techniques to capture a detailed scene of the farmland. This information would enable automated farm equipment guided by a 3D vision navigation system to perform automated weed control, spot spraying, and inter-row and inter-plant cultivation at a higher efficiency.

Computer Vision on the Road

Researchers at the Bundewehr University in Munich, Germany, are advancing their research on computer vision in road vehicles by equipping a Mercedes 500 SEL passenger car with a sense of vision. The autonomous road vehicle guidance system for normal traffic conditions detects road conditions, objects and obstacles from both the front and rear of the automobile using 4D machine vision. Computer vision in cars would reduce accidents and improve gas efficiency.

Computer vision conferences held annually by major organizations and associations focusing on the technology introduce numerous research papers on other uses of computer vision and its related fields. They range from simplifying everyday tasks to improving existing technology to creating a while new computer vision system for an application. Ongoing research is conducted at universities around the globe and by private computer vision research labs and companies. The extensiveness of the research being conducted in computer vision is indicative of the great potential it holds in many diverse fields and applications.

Machine Vision

Machine vision is perhaps the most successful offshoot of computer vision. Used in industrial settings since the late 1970s, machine vision is typically employed in manufacturing to conduct automatic inspections and for sorting and material handling. In addition, machine vision is being used for robot guidance in industrial and manufacturing settings. The success of machine vision has led to its own offshoots, among them video analytics. Traditionally used in automotive manufacturing to inspect parts, and later in license plate recognition, video analytics is now being considered for security technology, particularly wireless applications in retail settings and outdoor environments.

Biological Vision

Biological vision takes the principles of computer vision and applies them to humans and other animals. While computer vision interprets the processes of hardware and software in artificial vision systems, biological vision uses similar practices to study the physiological processes of visual perception and the human consciousness.

Robotic Vision

Robotic vision applies computer vision to the field of service robotics. Built-in cameras and sensors allow robots to move about unmanned, able to detect obstacles they encounter and translate back data of what they are seeing. Computer vision has enabled robotic vision to advance its applications into a wide range of fields, including the military, aerospace and space exploration, law enforcement, nuclear and industrial. In addition, research is focusing on areas that include human-robot interfaces, tele-assistance services and multi-sensorial perception.

Augmented Reality Applications

An offshoot of computer vision and computer graphics, augmented reality provides a live view of a real-world situation and enhances some of its elements by computer-generated applications, such as graphics, sounds, or GPS. An example of augmented reality is when television networks broadcasting a live pro football game display a yellow line over the field to indicate the distance from a first down or goal.

In this decade, we will continue to see offshoots of computer vision as the potential of the science continues to be discovered. And the future holds many possibilities for other branches of computer vision as well.

The Ancestry of Computer Vision

Since computer vision encompasses the science and technology of artificial intelligence and machines that see, computer science is the starting point when tracing the relationship of computer vision to other sciences. Computer science itself is a broad discipline, focusing on the science and technology of processing information. It encompasses computation theory, architecture of computer systems, theories of language programming and computer applications, all of which are utilized in computer vision. But artificial intelligence, a branch of computer science, is the science in which computer vision relies heavily upon. It focuses on enabling computers to do things that generally require human intelligence. In order to accomplish this mission, the science of artificial intelligence, along with computer vision, studies such aspects as learning, language interpretation, perception, reasoning and problem-solving.

A close relative to computer vision would be physics, in particular the optics, image sensors and measures involved in computer vision. Computer vision relies on physics when applying theories to the behavior of energy, motion and force. Another close relative is neurobiology, which studies the brain and nervous systems of living beings. This science allows computer vision to mimic the natural vision system and other human and animal behaviors. Also closely related to computer vision is mathematics, specifically signal processing. This field gives computer vision the methods for non-linear and multi-variable signals.

When it comes to application, computer vision turns to the other side of its family and the areas related to integrating the methods of theory. These relatives include image processing and analysis, automation and machine vision. Other methods used in computer vision are related to the fields of geometry, statistics, pattern recognition and optimization. With its reliance on numerous fields of math and science, computer vision has proved invaluable in the medical, military, aeronautics, construction, manufacturing, food processing and industrial areas.

The related sciences and fields of computer vision enable it to handle numerous tasks encompassing recognition, motion, image restoration and scene reconstruction. Among the key applications using computer vision are industrial image processing to handle processes and quality controls; monitoring involving data collection and safety systems, as well as event recognition; robotics encompassing assembly and the use of autonomous systems; and document analysis covering handwritten and graphics recognition. These tasks are wide-ranging and are used by numerous industries in countless applications throughout the world.

At its most basic, the tasks of computer vision can be categorized into five types:

1. Constructing a 3D View – Perhaps the most common task of computer vision is producing a 3D image from a flat or 2D image. This depth capability has proven beneficial in the medical field, as well as the military. For example, computer vision enables radar to generate a map of the surrounding ocean and identify the location of enemy ships by distinguishing them in 3D view. Besides actual applications, computer vision’s 3D view construction is also enhancing research in many areas.

2. Surface and Object Recognition – Recognizing objects and registering surface types are two common tasks performed by computer vision. It does this by capturing information, such as dimensions, composition and unique features, and producing accurate images so the object or surface can be identified. Computer vision’s surface and object recognition capabilities also are used for detection and identification purposes, such as detecting abnormal cells in medical imaging or identifying a specific person through fingerprints.

3. Scene or Event Reconstruction – Reconstructing a scene, whether from live data or a video, is another key task performed by computer vision. By inputting data of one or more scene images, computer vision can reconstruct the scene in a 3D model. As part of the reconstruction, computer vision takes into consideration essential factors, such as texture, lighting, spatial relations, gradients, and other influences. This computer vision task is often relied upon by the motion picture and film industry.

4. Motion Analysis – Tasks that require estimation of motion are often performed by computer vision. It can be used to determine velocity between two distinct points or within a scene that involves multiple objects, such as vehicles and humans. Accident reconstruction scene specialists often use motion analysis to determine the speed of a vehicle.

5. Image Restoration – Computer vision is used to recreate an image of an object or a structure that has been altered due to “noise,” such as a motion blur, sensor interference, or improper storage. It can also be used to enhance details in an image. Likewise, computer vision uses algorithms to restore image data that may have been corrupted. Applications that use computer vision’s image restoration capabilities range from art and film restoration to photography and live video streaming.

- Electronics that are more intelligent and aware of their environments;
- Game consoles that can detect body movements and facial expressions and use them as an interface between the game and player;
- Automotive electrical components that prevent accidents;
- Remote control applications that make the need for physical remote controls for televisions, lighting, heating and air conditioning, and the like obsolete;
- Swimming pool safety systems that monitor anyone in the water;
- The ability to analyze immense libraries of video content from a home television to quickly find content of interest to the viewer.

Embedded Vision Technology Holds The Key

The advancement of practical embedded vision technology is the link to enhancing computer vision in electronics. Embedded vision technology combines computer vision and embedded systems technologies to enable machines to interpret their environment through visual means. Embedded systems are already in use in kitchen appliances, electronic devices and consumer electronics. Because of the adaptability of embedded systems technology and low-cost depth sensors, it now can be integrated easily into existing electronic products, rather than require new designs.

Computer Vision To Be The Next Wireless

Marketing and industry experts are expecting big things from computer vision electronic products. They are basing their expectations on the public’s positive response to products already on the market that utilize computer vision. Among them are Microsoft Kinect, which enables the Xbox 360 to “see” and for players to use the gaming system without hand-held controllers. According to Microsoft, eight million Kinect units were sold in the first two months the product hit the market, making it the fastest selling consumer electronic device in history.

Brier, who in addition to his role as president of Berkeley Design Technology helped organize the Embedded Vision Alliance, predicts the pairing of computer vision and embedded vision technologies will be to the electronics industry what wireless was to the computer and mobile industries. He pointed out that consumers will soon come to expect it in their electronics.

Various Functions of Computer Vision Systems

Many computer vision systems, while diverse for their working environment, have many functions in common. Typical features include pre-processing, image acquisition and extracting of image features. For example, all computer vision systems require some form of image sensor to produce a digital image. While this function is typical in computer vision systems, the methods may vary, ranging from light-sensitive cameras and ultra-sonic cameras to radar and tomography devices. Some computer vision systems also have more advanced functions like segmentation, detection and high-level processing to process shape, texture, or motion.

More Robust Computer Vision Systems On The Horizon

As advances are made in technology, scientists are applying them to functions and applications in computer vision. For example, new developments in imaging and embedded computing technology has led scientists to incorporate portable imagers into computer vision systems that can be connected via wireless networks. The result is a computer vision system for video surveillance or any other function that calls for activity to be noticed, processed and interpreted.

Technological advances also led a group of European researchers to develop HERMES, a cognitive computer vision system that consists of video cameras and software designed to recognize and predict human behavior, and translate it into ordinary human language. This type of application can be used for accident prevention, intelligent surveillance and psychology efforts. Artificial intelligence researchers are also investigating the use of computer vision systems to better interpret the natural movements of people. Scientists at Purdue University are focusing on heat diffusion within a computer vision system to better achieve 3D object recognition for complex shapes or when the shape of an object takes on another form. The system would scan an object’s surface and use algorithms to forecast how heat would be diffused throughout it. Since heat diffusion patters are predictable, the shape of the object can be determined regardless of its orientation. For example, a computer vision system of this type would be able to recognize the shape of a closed fist as a human hand.

More robust computer vision systems are on the horizon as French researchers make headway with their study on autonomic computing. This would allow for a computer vision system designed for one application to be easily applied to new applications in a totally different genre. Autonomic computing is being looked at in building computer vision systems that are less costly to install, perform self-monitoring functions and provide enhanced reliability.

The ID: A Labs study identified the most common forms of identity theft on minors. Topping the list are credit card identity fraud and wireless activity fraud. According to the study, 60 percent of child identity theft incidents involved identity thieves opening up a credit card account using the child’s personal information. Also in the majority of fraud occurrences are cell phone and other wireless provider accounts opened under a child’s name. As Tom Oscherwitz, chief privacy officer at ID Analytics points out, identity thieves are on the lookout for children’s personal information that is left unprotected because their personal data is less likely to be monitored for misuse and opportunities to obtain credit under a child’s name are wide open.

The study was based on activity over a 12-month period from April 2010 to March 2011 of the more than 172,523 children enrolled in the Consumer Notification Service provided by ID Analytics. Of the 600 possible cases of child identity theft, 55 percent were confirmed as actual identity fraud by credit card issuers and other service providers. Oscherwitz said the solution to child identity theft needs to be a comprehensive one. In order to fully protect a child’s name, date of birth, social security number and other personal identifying information, he recommends a layered approach to identity theft protection that encompasses consumer education, technological advances, enhanced business processes and legislative guidance.

The growing incidences of child identity theft has prompted the Federal Trade Commission (FTC) and the U.S. Department of Justice’s Office for Victims of Crime (OVC) to hold a workshop entitled, Stolen Futures: A Forum on Child ID Fraud. Experts presenting at the workshop referenced additional research that found that identity thieves target children 51 times more than adults. The reason, the experts explain, is that a child’s Social Security number is unused, making it easier for identity thieves to associate a different name and birth date to it.

Child advocates recommend that parents become more vigilant in protecting their child’s personal information. For example, asking questions about how a child’s social security number will be used, handled and disposed of is important. Caution, awareness and monitoring, the experts agree, are the best preventative measures.

Related

• TransUnion Overview
• Identity theft
• ID Theft Protection Services

Some of the benefits that TransUnion provides consumers relate to fraud prevention and background checks. If a consumer suspects that there are victims of identity theft, they can verify the transactions that have been made under their credit identity by requesting a report from TransUnion. In addition, a yearly free credit report may be obtained in order to check for any discrepancies or inaccuracies. This can help save a consumer difficulty when it comes to obtaining credit.

Other uses for a TransUnion credit report can be related to organizations that extend credit to businesses and consumers. If an individual wishes to purchase a home mortgage, for example, a credit history will need to be submitted. TransUnion provides that service. The same holds true for any major purchase that requires the acquiring of a loan.

The credit report that TransUnion provides is basically a detailed summary of credit activity that is obtained from different sources. Once the requesting party has provided proof of identity, a hard copy of the report may also be provided. This copy can also include a FICO score, which is a weighted average of the complete transactions that have been undertaken by the party that is being assessed.

The TransUnion report, itself, is divided into six main categories which will include:

• Creditor Contacts – a listing of outstanding and current creditors and their contact information.
• Personal Information – including name(s), location address and phone numbers. Marital status and children are also noted.
• Summary of Transactions – credit accounts itemizations within a given period of time.
• Outstanding Accounts – detailed credit area transactions involving; real estate, collectibles, revolving fund accounts, instalments, etc.
• Previous Inquiries – a list of “hard” inquiries made regarding the credit history. These are usually made in order to assess the credit worthiness of an individual or business.
• Other information – this can include negative financial judgements, fees and taxes owed and paid for by the individual.

It must be noted, however, that not all credit monitoring agencies are required to report to all three credit bureaus. So it’s possible for omissions to be present on the TransUnion credit report that can be found on one of the other two major credit monitoring services.

When Credit Repair Services Are Beneficial

Companies that offer credit repair services are most advantageous for people who fall into any of the following categories:

Busy Lifestyle: If you know your credit reports contain errors or inaccuracies but you don’t have the time to contact all three of the credit reporting agencies and follow the procedures in the dispute resolution process, then a credit repair company is the best route to take since they do all the work for you in repairing your credit.

Not Very Well Organized: Fixing your credit yourself takes good organizational skills. If you know you’re not the best in documenting phone or written contact, organizing files for each dispute, or keeping good records, then credit repair services might be a better solution since account management is organized for you, often times online.

Prefer Professional Guidance: If you’re unsure about how to begin disputing credit report mistakes, credit repair services offer step-by-step programs that streamline the process. If you’re not comfortable with how to investigate or dispute negative items on your credit report, how to read a credit report, or what your rights are under consumer protection laws, you might feel more at ease using a credit repair company.

Identity Theft Victim: If you are or were a victim of identity theft, credit repair services have tools to address fraudulent accounts. These are accounts that have been opened in your name without your knowledge or consent. Credit repair services can assist identity theft victims with the steps necessary to stop fraudulent information from being reported on your credit report.

Issues Credit Repair Services Can Help You With
Credit repair companies focus on questionable items that appear on your credit reports. They can have such items removed if it can be proven that the debt is not yours or the action taken is not justified. Among the credit report issues that credit repair services look into are:

• • Improper credit card limits and balances;
• • Late payment errors;
• • Inquiries made on duplicate or closed accounts;
• • One debt reported several times;
• • Inaccurate student loans;
• • Invalid judgments made against you;
• • Invalid liens, charge-offs, bankruptcies, or foreclosures.

Deciding Your Credit Repair Route

When deciding whether to enroll in a credit repair service or do the credit repair yourself, be honest about how much time and effort you’re willing to put into the process if you do decide to do it yourself. The professional services offered by a credit repair company are designed to assist you with the process by doing much of the dispute process for you, including identifying incorrect information, sending out dispute letters, submitting supporting documentation, following up on responses and checking that the item in question has been adequately resolved.

At the very least, you can request your free credit report on your own and review it for inaccuracies. When deciding whether you need credit repair services, weigh the extent of the errors in your credit report with the extent of the commitment you’ll need to make doing the dispute work on your own. Also take into consideration your willingness to learn about credit report dispute procedures. If the process seems too time-consuming or overwhelming for you to do on your own, then credit repair services might be a right fit for you.