| Patent application number | Description | Published |
|---|
| 20090252437 | Image Skew Detection Apparatus And Methods - Methods and apparatus for detecting skew in a document image, such as a check image, to produce a de-skewed image are described. One example method includes detecting one or more lines in the image and determining whether the one or more lines are reliable. Reliability of a line may be based on at least one of line length, straightness, and the presence of holes in the line. If one or more lines are reliable, the method may calculate a skew angle of the image based on the one or more reliable lines' orientations with respect to an orientation of the image. A comparison may also be made between lines detected in different regions of the check to determine if a difference between skew angles corresponding to each of the compared lines is lower than an error threshold. | 10-08-2009 |
| 20110215151 | Method and Apparatus for Correcting Decoding Errors in Machine-Readable Symbols - The present invention is a system and method for decoding an image of a bar code. Decoding the barcode includes tokenizing a plurality of pixels in the image of the barcode based upon a plurality of thresholds to form a first set of tokens. Decoding the barcode also includes re-tokenizing the plurality of pixels in the image of the barcode based upon the intensity of the pixels in the plurality of tokens, and the relativity intensity of neighboring tokens, to form a second set of tokens. | 09-08-2011 |
| 20110215152 | Method and Apparatus for Creating Pixel Tokens from Machine-Readable Symbols to Improve Decoding Accuracy in Low Resolution Images - The present invention is directed towards systems and methods for decoding a bar code. Decoding the bar code includes segmenting the bar code into black tokens, white tokens and gray boundary pixels. Decoding the bar code also includes calculating a boundary position of each token with sub-pixel positional accuracy using intensity values of pixels in a boundary region. | 09-08-2011 |
| 20110249891 | Ethnicity Classification Using Multiple Features - Frontal face images are classified into four categories such as Asian, Caucasian, African and others. A new representation of face appearance, named BITF (Block Intensity and Texture Feature), is employed as the discrimination feature. An ensemble of three component classifiers, each trained with a different number of BITF features as inputs, is designed to achieve a reliable classification. Further reliability is obtained by taking into consideration other secondary features to boost the classification performance. | 10-13-2011 |
| Patent application number | Description | Published |
|---|
| 20080285860 | STUDYING AESTHETICS IN PHOTOGRAPHIC IMAGES USING A COMPUTATIONAL APPROACH - The aesthetic quality of a picture is automatically inferred using visual content as a machine learning problem using, for example, a peer-rated, on-line photo sharing Website as data source. Certain visual features of images are extracted based on the intuition that they can discriminate between aesthetically pleasing and displeasing images. A one-dimensional support vector machine is used to identify features that have noticeable correlation with the community-based aesthetics ratings. Automated classifiers are constructed using the support vector machines and classification trees, with a simple feature selection heuristic being applied to eliminate irrelevant features. Linear regression on polynomial terms of the features is also applied to infer numerical aesthetics ratings. | 11-20-2008 |
| 20090083332 | TAGGING OVER TIME: REAL-WORLD IMAGE ANNOTATION BY LIGHTWEIGHT METALEARNING - A principled, probabilistic approach to meta-learning acts as a go-between for a ‘black-box’ image annotation system and its users. Inspired by inductive transfer, the approach harnesses available information, including the black-box model's performance, the image representations, and a semantic lexicon ontology. Being computationally ‘lightweight.’ the meta-learner efficiently re-trains over time, to improve and/or adapt to changes. The black-box annotation model is not required to be re-trained, allowing computationally intensive algorithms to be used. Both batch and online annotation settings are accommodated. A “tagging over time” approach produces progressively better annotation, significantly outperforming the black-box as well as the static form of the meta-learner, on real-world data. | 03-26-2009 |
| 20090204637 | REAL-TIME COMPUTERIZED ANNOTATION OF PICTURES - A computerized annotation method achieves real-time operation and better optimization properties while preserving the architectural advantages of the generative modeling approach. A novel clustering algorithm for objects is represented by discrete distributions, or bags of weighted vectors, thereby minimizing the total within cluster distance, a criterion used by the k-means algorithm. A new mixture modeling method, the hypothetical local mapping (HLM) method, is used to efficiently build a probability measure on the space of discrete distributions. Thus, in accord with the invention every image is characterized by a statistical distribution. The profiling model specifies a probability law for distributions directly. | 08-13-2009 |
