Thursday, November 17, 2011

PC-based Machine Vision Versus Smart Camera Systems

PC-based Machine Vision Versus Smart Camera Systems

are also relatively simple, meaning that advanced computer skills are not always necessary for implementing a smart camera system.

Smart cameras can be designed with “open embedded” processing technology that eliminates the need for certain peripheral devices, such as a frame-capture card or an external computer. This streamlined system reduces costs while retaining functional capacity and, if coupled with an advanced processor, can rival some PC systems for power. In addition, the open embedded format is free of interfaces, which are usually needed to link multiple components into an operational network. Open embedded smart cameras are considered standalone vision systems, as they can generally handle tasks with minimal reliance on secondary devices.

Many cameras are equipped with communication hardware (Ethernet, RS232, etc.) that can transmit data to mainframes or other storage devices for future use. If the camera has limited processing ability, transferring information can augment processing speed by shifting administrative duties to an external device. Smart Camera Disadvantages

Due to their simple, compact design, smart cameras usually have less processing capability than PC-based systems. This limits the range and number of tasks that a smart camera can handle. Since they have difficulty managing more sophisticated algorithms, complex images or operations requiring rapid analysis are typically outside their scope.

Smart cameras often use proprietary hardware, making replacement or modification of parts a challenging process. When coupled with a compressed design, this feature severely limits the degree to which smart cameras can be upgraded. For similar reasons, expanding a smart camera’s functional range to handle different or additional tasks can be problematic. PC-based System Advantages

As previously mentioned, PC-based vision systems generally have greater processing power and are capable of handling complex operations at relatively high speed. This broader range of capabilities also enables PC systems to compensate for unexpected variables in certain tasks. For example, products on an assembly line tend to accumulate slight variations over a period of time, sometimes measured in months. Because the deviations occur gradually, they may not be detectable by a system with limited processing power, but a sophisticated computer vision network can perceive and mark the change.

Unlike smart cameras, most PCs are upgradeable and can have components swapped with relative ease. This versatility makes a PC system highly customizable, as it can have newer or more application-specific hardware installed to specialize on a certain task, or have its general range of functions expanded. For instance, a PC-based vision system could initially be employed for identifying and measuring components on an assembly line. Its duties could then be extended through installation of a software and hardware package to control a robotic arm that would remove any flawed products from the line. PC-based System Disadvantages

Since the PC itself is typically devoted to image-processing, several peripheral components are often necessary for frame grabbing, data transfer, lighting, and sometimes storage. This multi-unit format can become bulky or overly complex, and usually requires interfaces for each component in the system. Integrating such a network into a manufacturing process or existing vision system can be a challenging task, and may require advanced computer knowledge to install.

The multi-component configuration can also be highly fragile. Since numerous devices must be operating simultaneously, a malfunctioning unit can negatively affect the rest of the system, or even render it nonfunctional until the error can be corrected. PCs themselves tend to be less durable than smart cameras, and show greater signs of wear over a shorter period of time. Vision System Applications

Product inspection and quality assurance are two of the most common uses for machine vision. It is often employed for detecting flaws in vehicle parts for the automotive industry; pills, packaging, or labels for pharmaceutical companies; and microchips for computer manufacturers. Vision systems can scan barcodes for a product run, or retinas for biometric recognition.

Before deciding on a machine vision system, it is important for a manufacturer to evaluate the differences between formats. The distinction between smart cameras and PC-based systems can be summarized as one of cost versus capability. Smart cameras tend to be cheaper, more simply designed, and easier to use, but with limited processing power and expandability. PC systems have greater power, speed, and versatility, but at a higher price and with a more complicated design. However, as technological advances continue to refine the existing options, smart cameras and PC-based systems might eventually share more similarities than differences.

No comments:

Post a Comment