Image analysis – Applications – Animal – plant – or food inspection
Reexamination Certificate
1999-10-28
2001-10-30
Mancuso, Joseph (Department: 2623)
Image analysis
Applications
Animal, plant, or food inspection
C382S141000, C382S170000, C382S199000, C382S203000, C348S089000
Reexamination Certificate
active
06310964
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to the field of determining the size of an object from computerized optical image scanning devices. More specifically, the invention is a trainable system and method relating to recognizing the size of bulk items using image processing.
BACKGROUND OF THE INVENTION
Image processing systems exist in the prior art for estimating the size of objects. Often these systems use binary images to perform this computation. One common method is the use of the zero-th order moment of the binary picture function, or simply put, the number of picture elements that lie within the binary segment.
If the size and orientation of the object in the image is known in advance, which is the case in inspection problems, for example, binary matched filters are commonly used. This method allows for determining with great precision whether a specific object of certain size and shape is present in the image at a certain location.
The prior art often performs these methods to verify that the target object in the image is indeed the object that is expected, and, possibly, to grade/classify the object according to the quality of its appearance relative to its zero order moment or matched filter. An alternative purpose could be to identify the target object by matching the target image object with a number of reference matched filters.
In this description, identifying or measuring the size of one or more objects is defined as determining, given a set of reference sizes, the reference size of the target object. Classifying or grading the size is defined as determining that the target object is of a certain class representing a range of sizes and/or that the size of the object is satisfactory. Here, one of the classes can be a “reject” class, meaning that the size of the target object is not one of the expected values. Verifying, on the other hand, is defined as determining that the target is known to be a certain size and simply verifying this to be true or false. Recognizing is defined as identifying, measuring, classifying, grading, and/or verifying.
A round object, in this description, is an object having every part of the surface or circumference equidistant from the center. Bulk items include any item that is sold in bulk in supermarkets, grocery stores, retail stores or hardware stores.
Examples include produce (fruits and vegetables), sugar, coffee beans, candy, nails, nuts, bolts, general hardware, parts, and package goods.
In image processing, a digital image is an analog image from a camera that is converted to a discrete representation by dividing the picture into a fixed number of locations called picture elements and quantizing the value of the image at those picture elements into a fixed number of values. The resulting digital image can be processed by a computer algorithm to develop other images or characteristics of these images. These images or characteristics can be stored in memory and/or used to determine information about the imaged object. A pixel is a picture element of a digital image.
Image processing and computer vision is the processing by a computer of a digital image to modify the image or to obtain from the image properties of the imaged objects such as object identity, location, size, etc.
A scene contains one or more objects that are of interest and the surroundings which also get imaged along with the objects. These surroundings are called the background. The background is usually further away from the camera than the object(s) of interest.
Segmenting (also called figure/ground separation) is separating a scene image into separate object and background images. Segmenting refers to identifying those image pixels that are contained in the image of the object versus those that belong to the image of the background. The segmented object image is then the collection of pixels that comprises the object in the original image of the complete scene. The area of a segmented object image is the number of pixels in the object image.
Illumination is the light that illuminates the scene and objects in it. Illumination of the whole scene directly determines the illumination of individual objects in the scene and therefore the reflected light of the objects received by imaging apparatus such as video camera.
Ambient illumination is illumination from any light source except the special lights used specifically for imaging an object. For example, ambient illumination is the illumination due to light sources occurring in the environment such as the sun outdoors and room lights indoors.
Glare or specular reflection is the high amount of light reflected off a shiny (specular, exhibiting mirror-like properties—possibly locally) object. The color of the glare is mostly that of the illuminating light (as opposed to the natural color of the object).
A feature of an image is defined as any property of the image which can be computationally extracted. Features typically have numerical values that can lie in a certain range, say, R
0
-R
1
. In prior art, histograms are computed over a whole image or windows (sub-images) in an image. A histogram of a feature of an image is a numerical representation of the distribution of feature values over the image or window. A histogram of a feature is developed by dividing the feature range, R
0
-R
1
, into M intervals (bins) and computing the feature for each image pixel. Simply counting how many image or window pixels fall in each bin gives the feature histogram.
Image features include, but are not limited to, features that are related to the size of the objects in the image. The simplest features related to size of an object are the object pixels. The boundary pixels, subsets of boundary pixels, and characteristics determined from subsets of boundary pixels are also image features related to object size.
U.S. Pat. No. 4,515,275 to Mills and Richert discloses an apparatus and method for processing fruit and the like, particularly for sorting as a function of variables including color, blemish, size and shape. The fruit is moving on a conveyer belt while being rotated and imaged by a line scanning diode array. The line scanning diode array is sufficiently long such that the scanning line is longer than the fruit item and gives information about the length of the fruit. The number of individual detector signals which reflect presence of the fruit contains information to determine the width of the fruit. These numbers are squared and summed, the result being a representation of fruit volume, a characteristic related to fruit size.
U.S. Pat. No. 5,020,675 to Cowlin et al. discloses an apparatus for sorting conveyed articles. Sorting of food products such as vegetables or fruit, is achieved in accordance with their size, weight and color, or the presence of defects on them. Size is determined by the combination of examining the leading and following trailing edge of a trace and the color count of each article on the conveyer. To this information, weight information from load cells can be added.
The use of a zero order moment of a binary thresholded image of an object is an effective method for identifying the size of an object in an image. Similarly, the use of matched binary filters is effective for verifying the size of a target object in the image. The use of multiple matched binary filters allows for classifying the size of an object. The reason is that under well controlled imaging conditions, good segmentations can be obtained which, in turn, allow for precise measurements with the above methods.
Both for moment and matching techniques to work for object size recognition, very precise segmentations of the object from the background are needed. Furthermore, for matched filtering techniques, the exact orientation of the object in the image has to be known a priori.
STATEMENT OF PROBLEMS WITH THE PRIOR ART
In summary, much of the prior art in the agricultural arena is concerned with classifying/grading produce items. This prior art can only classify/identify objects/products/produce if they pass a scanner one object at a
Bolle Rudolf Maarten
Connell Jonathan Hudson
Mohan Rakesh
Bali Vikkram
International Business Machines Corp.
Mancuso Joseph
McGuireWoods LLP
Percello LouisJ.
LandOfFree
Produce size recognition system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Produce size recognition system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Produce size recognition system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2614004