Image analysis – Applications – Mail processing
Reexamination Certificate
2001-01-04
2003-09-02
Patel, Jayanti K. (Department: 2621)
Image analysis
Applications
Mail processing
C209S584000, C705S406000
Reexamination Certificate
active
06614916
ABSTRACT:
TECHNICAL FIELD
The present invention relates to machine vision systems. More particularly, the present invention relates to a machine vision system and a method for triggering a machine vision system.
BACKGROUND ART
Machine vision is currently utilized throughout commercial industry in a wide variety of applications. For example, machine vision technology is used in semiconductor wafer fabrication, pharmaceutical manufacturing, food processing, mailpiece processing, circuit board assembly, and many other areas. A common attribute of machine vision systems is the utilization of some type of video camera. The camera is used to obtain an electronic image of a feature of interest. For example, in mailpiece processing, a video camera may be used to obtain an electronic image of a bar code printed on a mailpiece.
In order to obtain an image, the camera exposes an array of light-sensitive devices to the image. The devices convert the light reflected from the object being imaged, into an electrical signal. The electrical signal is converted into digital data, which is transferred from the camera into the memory of the vision system. The vision system performs image processing and analysis on the image. For example, if the image is that of a bar code, the vision system may decode the bar code.
One problem in conventional machine vision systems is determining when to begin to acquire the image of interest. Most vision systems have a trigger input that instructs the camera as to when to acquire the image. The timing of this trigger input is critical to correct acquisition of the image. If the feature of interest is not within the field of view of the camera when the trigger signal is input to the camera, the correct image will not be acquired.
The problem of determining when to trigger acquisition of an image is particularly acute in the mailpiece processing industry. As used herein, the phrase “mailpiece processing” refers to the processing of mailpieces, i.e., sheet articles, that are destined for a postal service mailstream. In mailpiece processing, individual mailpieces include machine-readable symbols printed on the paper media comprising the mailpiece. The machine-readable symbols contain data that is critical to mailpiece processing, such as customer account numbers, mailpiece set and sequence numbers, page numbers, and insert selection data. A variety of symbologies can be used to encode this information, such as Code 39, Interleaved 2 of 5, DataMatrix, DataGlyph, and optical character recognition (OCR). It is the job of the vision system to read the codes printed on the mailpiece and decode the information correctly to ensure that the mailpiece is assembled and finished correctly.
In order for correct decoding of symbols to occur in mailpiece processing, a camera image of the symbol must be acquired by the vision system. For this to be accomplished, the acquisition must occur at the specific instant of time when the symbol resides within the camera's field of view. In other words, the vision system must be triggered at just the right moment. Such triggering is difficult in mailpiece processing due to the high speed at which mailpieces pass sensors and the fact that symbols are printed at different locations on different mailpieces.
A conventional solution to triggering acquisition of an image in mailpiece processing is to use a sensor separate from the camera to trigger acquisition of the video image. For example, the sensor might be an optical sensor that is mounted on a mailpiece processing device, such as a sheet feeder. Typically, the camera is positioned at a predetermined distance from the sensor such that when the sensor detects the edge of a mailpiece, the symbol printed on the mailpiece is within the field of view of the camera. Accordingly, when the sensor detects the edge, the sensor sends a signal to the vision system, which triggers the video camera to acquire the image of the symbol. In order to account for different symbol locations on different mailpieces, the camera is typically mounted on a mechanical slide so that the relative position between the camera and the optical sensor can be varied according to different jobs.
The external sensor approach to machine vision system triggering in mailpiece processing has a number of disadvantages. For example, using an external optical sensor to trigger acquisition of a video image requires additional hardware and cabling to be connected to the mailpiece processing equipment. An external optical sensor includes a housing and cabling that must be mechanically coupled to the mailpiece processing equipment. Because space on mailpiece processing devices, such as cutters and sheet feeders, is limited, finding a place to mount such additional hardware can be difficult. Moreover, additional hardware and cabling increases the cost of the machine vision system.
Another disadvantage associated with external sensors is that they can be unreliable in that they may trigger on features of an object other than the desired triggering feature. For example, in one mailpiece processing system, a reflective sensor was supposed to trigger on the edge of a mailpiece. The mailpiece at issue included dark horizontal lines printed parallel to the edge. The reflective sensor triggered on the edge and on the dark horizontal lines. Such false triggering caused the video camera to acquire incorrect images.
In another example, a mailpiece processing device, such as a cutter, may supply the triggering signal for the camera. A cutter is a mailpiece processing device that receives a continuous roll of paper and cuts the paper into sheets at predetermined intervals. Paper is fed from the roll into the cutter, and a blade cuts the paper into sheets that are sized according to the particular mailpiece processing job. Because the blade cuts the paper at a specific time, the time of actuation of the blade can be used to trigger acquisition of an image of a feature of interest.
The use of the cutter blade actuation time to trigger image acquisition is disadvantageous because the video camera must always be located at a fixed distance from the cutting blade. This distance is defined by the location or distance between the edge of a mailpiece corresponding to the cut and a feature of interest. This position may not be available on the cutter due to mechanical obstructions. If the position is not available, the bar code or other feature of interest must be moved to a different location on the mailpiece. This is undesirable because it limits flexibility in mailpiece layout. Accordingly, there exists a long-felt need for methods and systems for triggering machine vision systems that avoid at least some of the difficulties associated with conventional triggering methods.
DISCLOSURE OF THE INVENTION
The present invention includes methods and systems for machine vision system triggering that eliminate the need for triggering hardware separate from the video camera used to acquire the image. In one example, a machine vision system triggering method utilizes a programmable acquisition video camera to produce the triggering signal and to acquire the image. A programmable acquisition video camera is a video camera that allows the size of the field of view of the acquired camera image to be dynamically varied. Because the field of view can be varied, the time required to acquire and process an image can be varied. For example, it takes longer to acquire and process an image comprising the total field of view of the camera than it does to acquire and process an image comprising a single video line. Because the time for the camera to acquire and process an image is often limited by the speed at which objects pass the camera, it may be desirable to use only a portion of the total field of view of the video camera when speed is the primary concern and to use the full field of view when full image acquisition is the primary concern.
One instance in which speed is the primary concern is in detecting signatures of features used to trigger acquisition of a full frame image
Bell & Howell Mail and Messaging Technologies Company
McDermott & Will & Emery
Patel Jayanti K.
LandOfFree
Machine vision system and triggering method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Machine vision system and triggering method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Machine vision system and triggering method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3016007