Image analysis – Image compression or coding – Pyramid – hierarchy – or tree structure
Reexamination Certificate
1998-06-26
2001-04-10
Couso, Jose L. (Department: 2621)
Image analysis
Image compression or coding
Pyramid, hierarchy, or tree structure
Reexamination Certificate
active
06215907
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates generally to data compression in data storage and data communication systems and, more particularly, to a recursive wavelet data compression technique for use in storing and communicating compressed data on-line in, for example, a process control network.
DESCRIPTION OF RELATED ART
Process plants such as chemical refinery plants and drug manufacturing plants typically include a large number of field devices that control and measure parameters or variables at various locations within a process. A field device may be a control device (such as a flow valve controller), a measurement device (such as a temperature gauge, pressure gauge, flow meter, etc.) and/or any other device that affects, displays, stores or determines a value or other data associated with a process. Until the past decade or so, field devices have been rather simple units that were either manually or electronically controlled to produce one or, at most, a couple of electrical readings which were then communicated to a user over-wire or via an attached gauge. Generally speaking, these devices used only analog signals to communicate limited information pertaining to the readings or measurements made thereby.
More recently, so-called “smart” field devices have been developed. A smart field device is a device that is capable of communicating with and providing a host, a controller and/or a management system associated with a process, detailed information acquired by or stored in the device. For example, some smart field devices are capable of transmitting an analog and/or digital signal indicating a process value associated with the device (such as a measurement value) while also being capable of storing and transmitting digital signals indicating detailed device-specific information, such as calibration, configuration, diagnostic, maintenance and/or process information. Smart devices may, for example, store and transmit the units in which the device is measuring, the maximum ranges of the device, whether the device is operating correctly, troubleshooting information about the device, how and when to calibrate the device, etc. Furthermore, a smart field device may be able to perform operations on itself, such as self-tests and self-calibration routines. Exemplary smart devices include devices that follow the HART® (Highway Addressable Remote Transducer) protocol (HART devices), the FOUNDATION™ Fieldbus protocol (Fieldbus devices), the PROFIBUS® protocol, the WORLDFIP® protocol, the Device-Net® protocol, and the CAN protocol. However, other smart device protocols exist or may be developed in the future to support different types of smart devices.
Within standard process control networks, smart devices, as well as other types of field devices such as host devices, which may be controllers, data storage devices, user interface devices, etc., may be connected to a process controller via a dedicated line associated with each device or, alternatively, via a bus which interconnects all or at least a portion of the devices. Other process control networks, such as so-called distributed process control networks (in which control functions are performed by numerous control elements spread throughout the process control network) may have one or more buses interconnecting the devices needed to effect process control. Although many communication protocols, like those identified above, have been developed for process control communication networks, these control networks typically rely on only one (or a very limited number) of buses for all of the communication activities. In most systems, especially in larger systems, this fact places a premium on optimal usage of the bus. In fact, the more devices connected to the bus and the more information or data that is sent over the bus by the devices connected thereto, the more likely it is that bottlenecks will occur on the bus. These bottlenecks can result in lost data and reduced performance of the process control network. Furthermore, the increase in data communication capabilities of process control networks often taxes the ability of data storage devices, such as data libraries (data historians), to collect and store the data being sent over a bus. A recent move in the industry to use the process control bus to transmit video images, such as those taken by a video camera located somewhere within the process network, adds a significant load to the data traffic on the process bus and exacerbates the problem.
As is evident, it is desirable to configure a process control system (or any other communication network) to be capable of communicating as much data as possible using a given amount of bus bandwidth. Likewise, it is desirable to store as much data as possible using as little memory as possible in, for example, process control networks. In order to reach these goals, prior process control systems have used data compression techniques to compress data being stored in a storage device. However, most prior art data compression techniques used in process control systems have not been entirely satisfactory in reducing the amount of storage space required to store a given amount of data. Furthermore, prior art process control networks have not been able to reduce the bus bandwidth needed to communicate a given amount of data over a bus because they have not been capable of communicating compressed data via the bus.
Generally speaking, prior art process control networks that have compressed data for the purpose of data storage have used interoperative compression techniques, like the box car compression technique and the backward slope interpolation technique. Interpolative compression techniques typically use lower order interpolation such as zero order interpolation (e.g., the box car method) or first order interpolation (e.g., the backward slope method) over the current point and the last few recorded points to decide whether it is useful to store the current point. While these methods are fast and, therefore, may be used on-line or in real-time to compress and store incoming data, the decisions made by these methods are based on only a few points and, thus, these methods provide relatively poor performance.
Wavelet data compression techniques, on the other hand, have been shown to perform better than interpolative compression techniques with respect to various criteria like mean, square error, local point error, empirical transfer function ratio and cumulative scale error, mostly because wavelet data compression techniques take a greater number of data points into account when compressing data. However, wavelet data compression methods require a large set or window of data, such as all of the data corresponding to a particular video image or some other physical entity or logical unit, to be available in memory before compression can begin. As a result, known wavelet data compression techniques which have been applied in prior art data compression systems are relatively slow and inefficient because they sit inactive for a period of time while collecting an entire window of data and then must operate at high data processing speeds to compute the wavelet coefficients necessary to perform wavelet data compression. Likewise, wavelet data compression techniques require a relatively large amount of storage space to be set aside for uncompressed data during the compression process. Both of these factors make prior art wavelet data compression techniques unsuitable for use in real-time or on-line data compression wherein it is desirable to compress the incoming data efficiently in real-time while storing only a minimal amount of uncompressed data points.
SUMMARY OF THE INVENTION
The present invention is directed to a real-time or on-line data compression technique that has the speed, efficiency and on-line implementability of known interpolative compression techniques while maintaining the minimal storage qualities and accuracy or performance qualities of known wavelet data compression techniques. The compression technique of the pres
Blevins Terrence L.
Kumar Shailesh
Misra Manish
Qin S. Joe
Seemann Richard C.
Couso Jose L.
Fisher-Rosemont Systems, Inc.
Marshall O'Toole Gerstein Murray & Borun
LandOfFree
Recursive on-line wavelet data compression technique for use... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Recursive on-line wavelet data compression technique for use..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Recursive on-line wavelet data compression technique for use... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2443541