Smart cameras have been used in industrial applications for roughly two and a half decades, but advances in processor technologies have made the devices much more accessible and popular within the past 7 years, especially in areas such as machine vision and surveillance. However, when the term smart camera is mentioned, a wide variety of ideas still come to mind among individuals because there is no widespread agreement upon the definition of what a smart camera technically is. It is generally agreed upon that the basics of a smart camera include not only the image sensor, but also some type of processing chip: a CPU, DSP, FPGA, or other type of processing device (see Figure 1).
Today, however, even an off-the-shelf, point & shoot digital camera has some type of built-in image processing to remove the red eye effect, conduct facial recognition, apply a filter, or perform another type of image processing. So, if the inclusion of a processor along with an image sensor is not the defining attribute of a smart camera, what makes a smart camera “smart?” The key lies in the output.
Unlike most cameras, the primary output of a smart camera is not an image, but a decision or information. Since the image processing or machine vision algorithm is done directly on the smart camera, the image does not need to be passed onto a PC or another device. Instead, the result of the processing can be passed directly to an operator or another device in the system. For example, a smart camera may be selected for use in an in-line inspection system for a manufacturing line. The output of the smart camera could be a pass/fail report over a network to a database, a digital signal triggering a sorting system, or a serial command to a programmable logic controller (PLC).
A smart camera is a decision maker. Still, if you conduct an internet search for a smart camera, you will receive a large variety of results with very different features and appearance options. Let’s review some essential properties of smart cameras and how they have progressed:
As previously mentioned, the growing popularity of smart cameras can mainly be attributed to the increase in processor performance over the past decade. A 1 MHz smart camera 15 years ago would have been four times the size and cost of a >1 GHz smart camera of today. Many may be surprised when they realize the processing performance on some smart cameras rival what can be done on PC-based systems. An in-line inspection application, for example, that requires barcode reading, multiple geometric pattern matching, color analysis, line detection, and particle analysis can be done with a yield of over 25 parts per second. Additionally, this type of application can include triggering, lighting control, communication with other devices, and display all within the same compact package. Smart cameras come with a range of available processors, including DSPs, PowerPC-class, and Atom-class. There are also options with a mixed offering, such as a CPU with a DSP co-processor for certain algorithms.
One benefit to using a smart camera is that multiple components of a vision system are integrated into a single package, resulting in a small size and the potential to save a lot of space. The top-of-the-line, high-performance smart cameras keep getting faster without having to affect size, but the devices are also heading further into the low-end market. Reducing the functionality from an all-purpose smart camera to one that is designed for specific algorithms, such as barcode verification/reading and optical character recognition (OCR), reduces the number of components and complexity. With sizes smaller than 55 x 50 mm and weights of less than 60 grams, these types of smart cameras have been significantly growing in market availability.
Of course, a smart camera is still a camera and must acquire images. Both CMOS and CCD sensors can be found in smart cameras with resolutions of 5 MP, available in both color and monochrome. Smart cameras are not just limited to area scan anymore. Line-scan smart cameras are also available with frequencies over 10 kHz. While smart cameras do not cover the full range of options as normal cameras, they feature some of the most popular sensors.
A key downfall with smart cameras, however, comes when an application requires multiple image sensors connected to the same processing unit. One way to solve the problem is to synchronize multiple smart cameras, but that can add significant cost and complexity. In these cases, it is best to explore alternatives, such as compact vision systems or PC-based systems.
Most smart cameras today come with relatively simple-to-use software. Advanced knowledge of programming is not required to use the technology, but it is important to keep flexibility and scalability in mind. The investment to learn a new piece of software and write an application should be a somewhat long-term one. That is, the software should scale with application requirements and future projects.
Sensor and processor technologies are advancing rapidly, so the best-case scenario takes place when the smart camera model and software are well integrated, but not exclusive to each other. As a result, if you change smart camera models to a new version or need to move to a different hardware platform, such as a PC or operating system, a complete rewrite of the application or IP should not be required. The operating system running on the camera itself can also be a critical factor as Windows-based smart cameras connected to a plant network may fall under IT restrictions (see Figure 2).