The present invention relates to a method to generate a slap/fingers foreground mask to be used for subsequent image processing of fingerprints on an image acquired using a contactless fingerprint reader having at least a flash light, said method comprising the following steps: acquisition of two images of the slap/fingers in a contactless position in vicinity of the reader, one image taken with flash light on and one image taken without flash light, calculation of a difference map between the image acquired with flash light and the image acquired without flash light, calculation of an adaptive binarization threshold for each pixel of the image, the threshold for each pixel being the corresponding value in the difference map, to which is subtracted this corresponding value multiplied by a corresponding flashlight compensation factor value determined in a flashlight compensation factor map using an image of a non-reflective blank target acquired with flash light and to which is added this corresponding value multiplied by a corresponding background enhancement factor value determined in a background enhancement factor map using the image acquired without flash light, binarization of the difference map by attributing a first value to pixels where the adaptive binarization threshold value is higher than the corresponding value in the difference map and a second value to pixels where the adaptive binarization threshold value is lower than the corresponding value in the difference map, the binarized image being the slap/fingers foreground mask.

Visibility of a license plate and color reproducibility of a vehicle body are improved in a monitoring camera. A vehicle body area detection unit detects a vehicle body area of a vehicle from an image signal. A license plate area detection unit detects a license plate area of the vehicle from the image signal. A vehicle body area image processing unit performs processing of the image signal corresponding to the detected vehicle body area. A license plate area image processing unit performs processing different from the processing of the image signal corresponding to the vehicle body area on the image signal corresponding to the detected license plate area. A synthesis unit synthesizes the processed image signal corresponding to the vehicle body area and the processed image signal corresponding to the license plate area.

Techniques related to computer vision tasks are discussed. Such techniques include applying a pretrained non-linear transform and pretrained details boosting factor to generate an enhanced image from an input image and reducing the bit depth of the enhanced image prior to applying a pretrained computer vision network to perform the computer vision task.

Visibility of a license plate and color reproducibility of a vehicle body are improved in a monitoring camera. A vehicle body area detection unit detects a vehicle body area of a vehicle from an image signal. A license plate area detection unit detects a license plate area of the vehicle from the image signal. A vehicle body area image processing unit performs processing of the image signal corresponding to the detected vehicle body area. A license plate area image processing unit performs processing different from the processing of the image signal corresponding to the vehicle body area on the image signal corresponding to the detected license plate area. A synthesis unit synthesizes the processed image signal corresponding to the vehicle body area and the processed image signal corresponding to the license plate area.

A calculation method using pixel-channel shuffle convolutional neural network is provided. In the method, an operating system receives original input data. The original input data is pre-processed by a pixel shuffle process to be separated into multiple groups in order to minimize dimension of the data. The multiple groups of data are then processed by a channel shuffle process so as to form multiple groups of new input data selected for convolution operation. The unselected data are abandoned. Therefore, the dimension of the input data can be much effectively minimized. A multiplier-accumulator of the operating system is used to execute convolution operation using a convolution kernel and the multiple new groups of input data. Multiple output data are then produced.

Systems and methods for training image processing neural networks by synthetic photorealistic indicia-bearing images. An example method comprises: generating an initial set of images, wherein each image of the initial set of images comprises a rendering of a text string; producing an augmented set of images by processing the initial set of images to introduce, into each image of the initial set of image, at least one simulated image defect; generating a training dataset comprising a plurality of pairs of images, wherein each pair of images comprises a first image selected from the initial set of images and a second image selected from the augmented set of images; and training, using the training dataset, a convolutional neural network for image processing.

An apparatus for recognizing an object using an image includes a depth map generator that generates a depth map using a feature map of the image based on a dilated convolutional neural network (DCNN) and an object recognition device that recognizes the object using the depth map generated by the depth map generator and the image.

A media processing system and device with improved power usage characteristics, improved audio functionality and improved media security is provided. Embodiments of the media processing system include an audio processing subsystem that operates independently of the host processor for long periods of time, allowing the host processor to enter a low power state while the audio data is being processed. Other aspects of the media processing system provide for enhanced audio effects such as mixing stored audio samples into real-time telephone audio. Still other aspects of the media processing system provide for improved media security due to the isolation of decrypted audio data from the host processor.

This application provides an image recognition method, a storage medium, and a computer device. The method includes: obtaining a to-be-recognized image; preprocessing the to-be-recognized image, to obtain a preprocessed image; obtaining, through a first submodel in a machine learning model, a first image feature corresponding to the to-be-recognized image, and obtaining, through a second submodel in the machine learning model, a second image feature corresponding to the preprocessed image; and determining, according to the first image feature and the second image feature, a first probability that the to-be-recognized image belongs to a classification category corresponding to the machine learning model. It may be seen that, the solutions provided by this application can improve recognition efficiency and accuracy.

Digital image search training techniques and machine-learning architectures are described. In one example, a query digital image is received by service provider system, which is then used to select at least one positive sample digital image, e.g., having a same product ID. A plurality of negative sample digital images is also selected by the service provider system based on the query digital image, e.g., having different product IDs. The at least one positive sample digital image and the plurality of negative samples are then aggregated by the service provider system into a single aggregated digital image. At least one neural network is then trained by the service provider system using a loss function based on a feature comparison between the query digital image and samples from the aggregated digital image in a single pass.