Data processing: generic control systems or specific application – Specific application – apparatus or process – Article handling
Reexamination Certificate
2000-10-23
2002-11-19
Valenza, Joseph E. (Department: 3651)
Data processing: generic control systems or specific application
Specific application, apparatus or process
Article handling
C700S218000, C356S602000, C198S502200
Reexamination Certificate
active
06484066
ABSTRACT:
FIELD OF THE INVENTION
This invention relates generally to the conveying and sorting art and, more particularly, to the imaging of all sides of an object on a conveyor belt as the object passes by a camera for further processing of the object, such as reading an address label.
BACKGROUND OF THE INVENTION
Machine vision can replace human vision for a variety of tasks such as inspection and target tracking and can do a better job than human vision when precise information must be quickly and/or repetitively extracted from an image. An example of a task where machine vision inspection is needed is package handling for a package delivery service. A delivery service such as the U.S. Postal Service typically handles millions of items each year. Package handling requires some automation, such as using machine vision to read addresses, a necessity in order to deliver all the items to their proper destinations quickly.
Machine vision inspection of moving objects can be accomplished using a line scan camera. A line scan camera has a linear array of photosensors and can generate a two-dimensional image of an object as the object moves past the camera.
U.S. Pat. No. 5,249,035 discloses a method of measuring three dimensional shape objects using multiple cameras to gather images.
U.S. Pat. No. 5,543,103 discloses a process of surface shaping by creating a three dimensional depiction of an object and then incorporating a selected portion as an integral part of a wall surface.
U.S. Pat. No. 5,852,672 discloses an image system for three dimensional, 360 degree, time sequence surface mapping of moving objects.
U.S. Pat. No. 5,877,963 discloses an intelligent document recognition and handling arrangement for analyzing an image and comparing it to a data base of previously acquired images.
U.S. Pat. No. 5,864,640 discloses a method and apparatus for optically scanning three dimensional objects using color information in trackable patches and is concerned with using color information from a series of two dimensional color images to derive a three dimensional location in space of the surface points which produced the color images.
U.S. Pat. No. 5,838,573 discloses a process and apparatus for spotting labels using a camera for each side of an object and the images are used for the determination of the coordinates so that wraps on one conveyor may be mated with the containers which are located on another conveyor.
One problem presented in automatic inspection systems is the inability to image all sides of a package on a conveyor belt as the package passes by a camera. The top and bottom of the package can be imaged using current technology. The two sides of the package parallel to the direction of travel may also be imaged using current technology if the package is oriented squarely on the conveyor belt. Using a line scan camera, however, which is the economical method of imaging moving objects on a conveyor system, it is very difficult to obtain a high quality of image of those sides of the package that change their distance from the camera as the package passes through the camera viewing area. In the case of a package that is registered squarely on the conveyor belt, the leading and trailing sides would significantly change focal distance of the camera's scan line as it sweeps across the surface. It is even more difficult with an unregistered (i.e. skewed) package because none of the four vertical sides of the package maintain a constant focal distance as the package passes by the camera.
A device that is currently used in industry is a bar code reader tunnel scanner. This device will read bar codes on any side of a package on a conveyor belt as the package passes by the camera. A bar code reader, however, only examines the pulse stream of a reflected laser beam and looks for the specific pulse pattern produced when the beam sweeps over a bar code. It is not possible to use this information to build an image of the surface being viewed that is of adequate quality for an optical character recognition (OCR) process.
The shortcomings of current imaging technology require either a manual operation that orients the package so that the camera views the side of interest, or a complicated mechanism that reorients the package until all the sides have been scanned. The manual operation is slow and has variable reliability and added expense of human intervention. The reorientation mechanism is a more consistent capital expense than the manual operation, but is complicated due to the range of size, weight, shape and texture of the packages in the mail stream, which it must be capable processing.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a way of imaging all sides of the package quickly and accurately.
It is another object of the present invention to provide a method and system capable of acquiring images from all visible exterior surfaces of an object, for example, in a machine vision inspection operation.
It is a further object of the present invention to provide a method and system which is capable of incorporating therein extended depth of field technology during machine vision inspection.
These objects as well as others objects and advantages of the present invention are achieved by the embodiments of the invention.
The present invention uses cameras with extended depths of field to view objects angularly so that at least two sides of a six sided box, for example, can be viewed by a single camera, and in some cases three sides may be scanned by a single camera. For example, the top and a side may be scanned simultaneously with the near and far portions of each surface remaining in focus. In this case the camera scans a line from the bottom of a side to the top of the side and then bends over onto the top of the box and extends to the far edge on the top of the box.
Typically there are scan of 200 of these lines for every inch that the box moves past the camera on the conveyer. The result of placing these scanned lines side by side is an image of two sides of the box with every portion of each surface being in focus. Alternatively the front and then the side would be scanned with vertical scan lines as the box moved past the camera. Thus, “tunnel scanning” may be used to collect images for various purposes, such as, for an OCR system. A conventional camera would have to continually change focus, as the parcel passes the camera on a conveyor in order to capture an image of the front of the box as it approached or of the back of the box as it moved away from the camera. If the box is not aligned to the camera's line of sight, then even refocusing would not allow the entire scan line to be in focus.
The tunnel scanning system, using the extended depth of field technology can provide images of all six sides of a box on a conveyor passing a camera with as few as two line scanning cameras, each collecting images of three sides of the box.
An extended depth of field imaging system is used in an image lift tunnel scanner capable of simultaneously focusing on objects of different heights. A typical optical system, in inspection applications, has a limited depth of focus of about +/−1 inch, and must adjust its focus to view objects of more widely varying heights separately. The extended depth of field imaging system is capable of simultaneously viewing packages that have as much as a 30 inch height differential. The application places a camera above the conveyor belt, where the camera scans a mixed stream of non-singulated packages. Short and tall packages that are side by side are scanned with the same focal quality.
The image lift tunnel scanner operates by passing a package first through a dimensioning device, such as light curtain, that determines the package's height, width, length, orientation and position on the conveyor belt. A bottom scan camera takes a fixed focus image and requires only a trigger and length information from the light curtain. The length information determines the duration of the scan after the camera receives the trigger signal
Albertelli Lawrence E.
Riess Michael J.
Crawford Gene O
Erlich Jacob N.
Kaye Harvey
Lockheed Martin Corporation
Perkins Smith & Cohen LLP
LandOfFree
Image life tunnel scanner inspection system using extended... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image life tunnel scanner inspection system using extended..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image life tunnel scanner inspection system using extended... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2993446