Abstract:
A banknote recognition method based on sorter dust accumulation and a sorter. An effective region boundary is determined by using a gray-scale difference between a foreground and a background of a sensor image signal, an edge is searched for by comprehensive means of signal features of various sensors, detection direction change and secondary scanning, and finally the effective boundary of the image region is relocated, so that the detection rate and recognizing accuracy of the sorter can be greatly improved. The sorter comprises a banknote inlet, a banknote outlet, a banknote exit port, a conveying rail and a recognizing module. The recognizing module comprises two sets of CIS image sensors arranged oppositely, two sets of light transmitting plates arranged oppositely, a storage module, a detection module and a display module.
Abstract:
An image calibration method and device. A deviation compensation coefficient of each pixel is calculated according to image data generated by strong and weak lighting parameters, and the deviation compensation coefficient is used to calibrate data of each image pixel collected and output by an image sensing unit. Therefore, the influence of inconsistency of each photosensitive element of the image sensing unit on the image greyscale can be eliminated, so that the effect of document medium image recognition is increased. The method in the embodiments of the present invention comprises: S1, acquiring a strong lighting parameter of each photosensitive element within a long exposure time and a weak lighting parameter thereof within a short exposure time; S2, acquiring image data of a sheet medium; and S3, according to the strong lighting parameter, the weak lighting parameter and the image data, calculating the calibrated image data of the sheet medium.
Abstract:
A complex background-oriented optical character recognition method and device are provided. The method of the present invention includes: collecting image information to obtain a collected image; according to character characteristics, acquiring a target character region from the collected image, and taking same as a target object; extracting character edge information in the target object using a differential method to obtain an extracted image; superposing the target object and the extracted image to obtain a recovery image; conducting inversion and Gaussian filtration processing on the recovery image to obtain a processed image; searching for a target character location in the processed image; and recognizing the target character location. On this basis, accurate and quick locating and recognition of characters can be realized on the basis of effectively suppressing background noise and highlighting character information.
Abstract:
An anti-counterfeiting feature generation method for a valuable document and an authentication method and device therefor. The anti-counterfeiting feature generation method for a valuable document uses the anti-counterfeiting feature information redundancy to hide accurate information about an anti-counterfeiting feature. In the generation method, the anti-counterfeiting feature of a valuable document has little change on the human perception, but a valuable document authentication device in a financial self-service equipment can effectively extract the hidden accurate information about an anti-counterfeiting feature from signals obtained by a sensor and conduct quantitative detection and authentication, thereby effectively authenticating whether the valuable document is counterfeit or not. A new anti-counterfeiting feature which facilitates machine recognition is loaded to the valuable document, and a corresponding valuable document authentication device is configured to a financial self-service equipment, so that the financial self-service equipment can read the anti-counterfeiting feature of the valuable document by facilitating machine recognition, thereby improving the authentication accuracy of the financial self-service equipment to the valuable document.
Abstract:
A paper medium skew correcting device includes: a support including a pair of fixed stop plates arranged in parallel; a skew correcting tapered roller assembly including a tapered roller, a reference plate located at a large end face of the tapered roller and a transmission shaft; a tapered cover plate covering on the tapered roller with a skew correcting passage formed between an inner wall of the tapered cover plate and an outer wall of the tapered roller, an auxiliary pressure roller mechanism is mounted on an outer wall of the tapered cover plate and is configured to press a paper medium against the outer wall of the tapered roller tightly, the tapered cover plate is provided with an opening; and a driving transmission assembly. The device fully uses combined action of centrifugal force and friction force to realize skew correction and conveying of the paper medium.
Abstract:
A method and device for banknote identification based on thickness signal identification, allowing for simple and highly efficient identification and separation of abnormal banknotes by means of combining two thickness identification methods, namely an upward-facing area identification method and a downward-facing area identification method, for banknote identification. The recognition method comprises: collecting a thickness signal of a banknote; preprocessing the thickness signal; utilizing an upward-facing area identification method to identify the thickness signal, thus acquiring an upward-facing processing identification result; utilizing a downward-facing area identification method to identify the thickness signal, thus acquiring a downward-facing processing identification result; merging the upward-facing identification processing result and the downward-facing processing identification result on the basis of a predetermined rule, acquiring a merged result; and, identifying the merged result, thus acquiring an identification result.
Abstract:
The present invention relates to a recognition method and a recognition device for a sheet-type medium. The method comprises: step 1, collecting image information; step 2, splitting from the image information collected image information of an object-to-be-recognized; step 3, reconstructing image data of a feature area, extracting a feature on the reconstructed image data, and forming an eigenvector; and, step 4, sending the eigenvector into a standard data-trained classifier, and acquiring a recognition result. The method has an image reconstructed on the basis of feature area prior information, improves image quality, facilitates increased recognition accuracy, and, because only the image area that requires feature extraction is reconstructed, reduces storage space consumption. This not only saves large amounts of storage space and time and increases resource utilization rate, but also increases recognition accuracy and achieves an improved balance between resource and efficiency.
Abstract:
Provided are a method and a device for offline identity authentication. The method includes: acquiring two or more images for identity authentication to constitute a to-be-authenticated multivariate image group; extracting a concatenated PCA convolution feature of each of the images in the to-be-authenticated multivariate image group, to obtain feature vectors; fusing information on the images in the to-be-authenticated multivariate image group based on the feature vectors according to a score fusion strategy with a supervisory signal, to obtain a fusion vector; and inputting the fusion vector to a pre-trained SVM classifier to authenticate and determine whether the images in the to-be-authenticated multivariate image group are consistent with one another, to obtain an identity authentication result.
Abstract:
A method and device for banknote identification based on thickness signal identification, allowing for simple and highly efficient identification and separation of abnormal banknotes by means of combining two thickness identification methods, namely an upward-facing area identification method and a downward-facing area identification method, for banknote identification. The recognition method comprises: collecting a thickness signal of a banknote; preprocessing the thickness signal; utilizing an upward-facing area identification method to identify the thickness signal, thus acquiring an upward-facing processing identification result; utilizing a downward-facing area identification method to identify the thickness signal, thus acquiring a downward-facing processing identification result; merging the upward-facing identification processing result and the downward-facing processing identification result on the basis of a predetermined rule, acquiring a merged result; and, identifying the merged result, thus acquiring an identification result.
Abstract:
An anti-counterfeiting feature generation method for a valuable document and an authentication method and device therefor. The anti-counterfeiting feature generation method for a valuable document uses the anti-counterfeiting feature information redundancy to hide accurate information about an anti-counterfeiting feature. In the generation method, the anti-counterfeiting feature of a valuable document has little change on the human perception, but a valuable document authentication device in a financial self-service equipment can effectively extract the hidden accurate information about an anti-counterfeiting feature from signals obtained by a sensor and conduct quantitative detection and authentication, thereby effectively authenticating whether the valuable document is counterfeit or not. A new anti-counterfeiting feature which facilitates machine recognition is loaded to the valuable document, and a corresponding valuable document authentication device is configured to a financial self-service equipment, so that the financial self-service equipment can read the anti-counterfeiting feature of the valuable document by facilitating machine recognition, thereby improving the authentication accuracy of the financial self-service equipment to the valuable document.