Vehicle interior classification system and method
First Claim
1. A method for determining a classification of an object of interest relating to a vehicle, comprising the steps of:
- (i) with an image capture device disposed at the vehicle and having a field of view, acquiring an image;
(ii) limiting the dynamic range of the acquired image to obtain a resultant image that comprises at least one region of interest, which encompasses a region of the field of view that is less than the field of view of the image capture device;
(iii) processing image data representative of said at least one region of interest to extract information relating to a set of image features from the acquired resultant image while not processing image data outside of said at least one region of interest;
(iv) statistically processing the extracted information with a previously determined set of image feature information values, each member of the set of image feature information values corresponding to a respective one of a set of predefined classifications of an object of interest relating to the vehicle, to determine the most probable classification;
(v) outputting the determined most probable classification of an object of interest related to the vehicle; and
wherein step (iii) comprises processing image data of the resultant image with a two dimensional complex discrete wavelet transform to produce a selected set of coefficients related to features in the acquired resultant image and, in step (iv), the previously determined set of image feature values comprises a set of weights for each defined classification, the weights being multiplied with the set of coefficients to produce a score for each defined classification, the defined classification with the highest produced score being the classification output in step (v).
2 Assignments
0 Petitions
Accused Products
Abstract
A vehicle interior classification system and method determines a classification relating to the interior of the vehicle, such as the occupancy status of a vehicle seat or the state of alertness of a vehicle driver, from one or more images of an appropriate portion of the interior of the vehicle acquired with an image capture device. The acquired images may be processed to limit the dynamic range of the images to obtain a resultant image that includes one or more regions of interest which are less than the total field of view of the image capture device. The resultant images are processed to extract information about features in the image. The set of coefficients produced with suchprocessing can be reduced to a subset of the total number of coefficients, the members of the subset being selected for their ability to discriminate between the classifications defined for the system.
24 Citations
17 Claims
-
1. A method for determining a classification of an object of interest relating to a vehicle, comprising the steps of:
-
(i) with an image capture device disposed at the vehicle and having a field of view, acquiring an image; (ii) limiting the dynamic range of the acquired image to obtain a resultant image that comprises at least one region of interest, which encompasses a region of the field of view that is less than the field of view of the image capture device; (iii) processing image data representative of said at least one region of interest to extract information relating to a set of image features from the acquired resultant image while not processing image data outside of said at least one region of interest; (iv) statistically processing the extracted information with a previously determined set of image feature information values, each member of the set of image feature information values corresponding to a respective one of a set of predefined classifications of an object of interest relating to the vehicle, to determine the most probable classification; (v) outputting the determined most probable classification of an object of interest related to the vehicle; and wherein step (iii) comprises processing image data of the resultant image with a two dimensional complex discrete wavelet transform to produce a selected set of coefficients related to features in the acquired resultant image and, in step (iv), the previously determined set of image feature values comprises a set of weights for each defined classification, the weights being multiplied with the set of coefficients to produce a score for each defined classification, the defined classification with the highest produced score being the classification output in step (v). - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A system for determining a classification of an object of interest relating a vehicle, comprising:
-
an image capture device disposed at a vehicle and having a field of view and operable to acquire an image; an image capture subsystem operable to process the image to limit the dynamic range of the image to obtain a resultant image that comprises at least one region of interest, which encompasses a region of the field of view that is less than the field of view of the image capture device; wherein said image capture subsystem processes image data of the resultant image with a two dimensional complex discrete wavelet transform to produce a selected set of coefficients related to features in the acquired resultant image; an image feature extractor operable to process image data representative of said at least one region of interest to extract information relating to a set of image features from the acquired resultant image while not processing image data outside of said at least one region of interest, in order to produce for the resultant image a set of values corresponding to features in the resultant image; a classifier operable to combine the set of values produced by the image feature extractor with a predetermined set of classification values, each classification value corresponding to a different possible predefined classification, the results of the combination representing the probability that each predefined classification is the current classification, the classifier operable to select and output the most probable classification; and wherein the results of the combination comprise a set of weights for each defined classification, the weights being multiplied with the set of coefficients to produce a score for each defined classification, the defined classification with the highest produced score being the most probable classification selected and output by the classifier. - View Dependent Claims (11, 12, 13, 14, 15, 16, 17)
-
Specification