The present invention discloses a method and system for splicing and restoring shredded paper based on an extreme learning machine. The method includes: acquiring a shredded paper training sample to be spliced; extracting left and right boundary feature data of the training sample; training an extreme learning machine neural network model according to the left and right boundary feature data, to obtain a trained neural network model; acquiring a shredded paper test sample to be spliced; extracting left and right boundary feature data of the test sample; selecting a first piece of to-be-spliced shredded paper; selecting shredded paper with a highest degree of coincidence with the first piece of to-be-spliced shredded paper by the trained neural network model; determining whether the shredded paper with the highest degree of coincidence is correctly spliced to the first piece of to-be-spliced shredded paper; if yes, splicing shredded paper until all the shredded paper is spliced and restored; and if not, adopting manual marking, and continuing to select shredded paper with a highest degree of coincidence with the first piece of to-be-spliced shredded paper by the trained neural network model. The method and system for splicing and restoring shredded paper based on an extreme learning machine can well splice and restore shredded paper quickly.

Disclosed is a method and system for splicing and restoring shredded paper based on an extreme learning machine (ELM). The method includes: acquiring a shredded paper training sample to be spliced; extracting left and right boundary feature data of the sample; training an ELM neural network model according to the feature data to obtain a trained neural network model (TNNM); acquiring a shredded paper test sample to be spliced; extracting feature data of the test sample; selecting a first piece of to-be-spliced shredded paper; selecting, by the TNNM, a shredded piece with a highest degree of coincidence with the first piece; determining whether the shredded piece is correctly spliced to the first piece; if yes, splicing shredded paper until all shredded paper is spliced and restored; if not, adopting manual marking, and continuing to select, by the TNNM, shredded paper with a highest degree of coincidence with the first piece.

Systems and methods for efficient implementation of a convolutional layer of a convolutional neural network are disclosed. In one aspect, weight values of kernels in a kernel stack of a convolutional layer can be reordered into a tile layout with tiles of runnels. Pixel values of input activation maps of the convolutional layer can be reordered into an interleaved layout comprising a plurality of clusters of input activation map pixels. The output activation maps can be determined using the clusters of the input activation map pixels and kernels tile by tile.

An enhanced encoder system generates residual bitstreams representing additional image information that can be used by an image enhancement system to improve a low quality image. The enhanced encoder system upsamples a low quality image and compares the upsampled image to a true high quality image to determine image inaccuracies that arise due to the upsampling process. The enhanced encoder system encodes the information describing the image inaccuracies using a trained encoder model as the residual bitstream. The image enhancement system upsamples the same low quality image to obtain a prediction of a high quality image that can include image inaccuracies. Given the residual bitstream, the image enhancement system decodes the residual bitstream using a trained decoder model and uses the additional image information to improve the predicted high quality image. The image enhancement system can provide an improved, high quality image for display.

The disclosure extends to methods, systems, and apparatuses for automated fixation generation and more particularly relates to generation of synthetic saliency maps. A method for generating saliency information includes receiving a first image and an indication of one or more sub-regions within the first image corresponding to one or more objects of interest. The method includes generating and storing a label image by creating an intermediate image having one or more random points. The random points have a first color in regions corresponding to the sub-regions and a remainder of the intermediate image having a second color. Generating and storing the label image further includes applying a Gaussian blur to the intermediate image.

A method of analyzing a fingerprint, the method comprising the step of acquiring a fingerprint image (20) together with the following steps: performing filtering processing on the fingerprint image to estimate, for each pixel of the fingerprint image, a first frequency of the ridges (21) in the fingerprint, and using the first frequencies associated with the pixels of the fingerprint image to produce a first frequency map (22) of the fingerprint image; subdividing the fingerprint image into a plurality of windows each comprising a plurality of pixels, calculating a Fourier transform for each window in order to estimate a second frequency of the ridges for all of the pixels in said window, and using the second frequencies associated with the pixels of the windows to produce a second frequency map of the fingerprint image; and merging the first frequency map and the second frequency map in order to obtain a map of consolidated frequencies of the fingerprint image.

A surface defect inspection method includes: acquiring an original image by capturing an image of a subject of an inspection; generating texture feature images by applying a filtering process using spatial filters to the original image; generating a feature vector at each position of the original image, by extracting a value at a corresponding position from each of the texture feature images, for each of the positions of the original image; generating an abnormality level image representing an abnormality level for each position of the original image, by calculating, for each of the feature vectors, an abnormality level in a multi-dimensional distribution formed by the feature vectors; and detecting a part having the abnormality level that is higher than a predetermined level in the abnormality level image as a defect portion or a defect candidate portion.

A method for encoding space-time signals comprises: collecting space-time signals of various local spatial positions in a monitoring area, and accumulating the space-time signals according to time, so as to obtain cumulative signal intensity values; transforming the cumulative signal intensity values by means of a filter, and outputting a pulse signal when a transformation result exceeds a specific threshold; arranging pulse signals corresponding to a local spatial position into a sequence according to the time, so as to obtain a pulse sequence expressing the local spatial position signals and a change process thereof; and arranging the pulse sequences of all local spatial positions into a pulse sequence array according to interrelation among the spatial positions to serve as an encoding for dynamic space-time signals of the monitoring area.

Embodiments of the present disclosure provide an expression recognition method, apparatus, electronic device and storage medium. An expression recognition model includes a convolutional neural network model, a fully connected network model and a bilinear network model. During an expression recognition process, after an image to be recognized is pre-processed to obtain a facial image and a key point coordinate vector, the facial image is computed by the convolutional neural network model to output a first feature vector, the key point coordinate vector is computed by the fully connected network model to output a second feature vector, the first feature vector and the second feature vector are computed by the bilinear network model to obtain second-order information, and an expression recognition result in turn is obtained according to the second-order information. During this process, robustness of gestures and illuminations is better, and accuracy of expression recognition is improved.

Apparatuses, methods, and systems are presented for reacting to scene-based occurrences. Such an apparatus may comprise dedicated computer vision (CV) computation hardware configured to receive sensor data from a sensor array comprising a plurality of sensor pixels and capable of computing one or more CV features using readings from neighboring sensor pixels of the sensor array. The apparatus may further comprise a first processing unit configured to control operation of the dedicated CV computation hardware. The first processing unit may be further configured to execute one or more application programs and, in conjunction with execution of the one or more application programs, communicate with at least one input/output (I/O) device controller, to effectuate an I/O operation in reaction to an event generated based on operations performed on the one or more computed CV features.

Methods, system, and media for comparing a set of images to determine the existence and location of any differences between the image set. The differences may be located using image comparison techniques such as SURF and Blob Detection, as well as through techniques used to identify areas of data sliding and match probabilities. A logical match probability, as well as a physical match probability, may be included in an output report with a result image highlighting the differences between the comparison images in the image set.