In one embodiment, an apparatus comprises a communication interface and a processor. The communication interface is to communicate with a plurality of devices. The processor is to: receive compressed data from a first device, wherein the compressed data is associated with visual data captured by sensor(s); perform a current stage of processing on the compressed data using a current CNN, wherein the current stage of processing corresponds to one of a plurality of processing stages associated with the visual data, and wherein the current CNN corresponds to one of a plurality of CNNs associated with the plurality of processing stages; obtain an output associated with the current stage of processing; determine, based on the output, whether processing associated with the visual data is complete; if the processing is complete, output a result associated with the visual data; if the processing is incomplete, transmit the compressed data to a second device.

The present invention relates to an image processing device and a method of framing changes and movements in a video image divided into N×N blocks of pixel positions. The method comprises calculating a first bitmap of the video image by a DCT transform on each of the N×N blocks of pixel positions, assigning a first binary value to the pixel positions of the N×N blocks when more than an amount of change, and a second binary value to the pixel positions of the N×N blocks when less than an amount of change. Calculating a third bitmap by an OR operation between a number of bitmaps representing past time frames of the video image, calculating a fourth bitmap by performing a dilation process of the third bitmap representing the current time frame of the video image, and creating one or more frames identifying area of changes and movements in the video image based on detecting BLOBs (Binary Large Objects) in the fourth bitmaps.

In one embodiment, an apparatus comprises a communication interface, a memory, and a processor. The communication interface is to communicate with one or more devices. The memory to store a device identity blockchain. The processor is to: receive a device identity transaction from a first device, wherein the device identity transaction comprises a device identity; compute a hash of the device identity; determine, based on the hash, whether the device identity is registered in the device identity blockchain; and upon a determination that the device identity is not registered in the device identity blockchain, add the device identity transaction to the device identity blockchain.

Metrology methods, modules and systems are provided, for using machine learning algorithms to improve the metrology accuracy and the overall process throughput. Methods comprise calculating training data concerning metrology metric(s) from initial metrology measurements, applying machine learning algorithm(s) to the calculated training data to derive an estimation model of the metrology metric(s), deriving measurement data from images of sites on received wafers, and using the estimation model to provide estimations of the metrology metric(s) with respect to the measurement data. While the training data may use two images per site, in operation a single image per site may suffice—reducing the measurement time to less than half the current measurement time. Moreover, confidence score(s) may be derived as an additional metrology and process control, and deep learning may be used to enhance the accuracy and/or speed of the metrology module.

In one embodiment, an apparatus comprises a storage device and a processor. The storage device stores a plurality of images captured by a camera. The processor: accesses visual data associated with an image captured by the camera; determines a tile size parameter for partitioning the visual data into a plurality of tiles; partitions the visual data into the plurality of tiles based on the tile size parameter, wherein the plurality of tiles corresponds to a plurality of regions within the image; compresses the plurality of tiles into a plurality of compressed tiles, wherein each tile is compressed independently; generates a tile-based representation of the image, wherein the tile-based representation comprises an array of the plurality of compressed tiles; and stores the tile-based representation of the image on the storage device.

Noise reduction in images is provided by performing a noise reduction step on blocks of pixels within a video-processing pipeline. The noise reduction step consists of applying a discrete cosine transform (DCT) to the block of pixels, quantizing the resulting DCT coefficients, and performing an inverse of the DCT to the quantized coefficients. The output of that noise reduction step is a block of image pixels similar to the input pixels, but with significantly less image noise. Because the noise reduction step can be performed quickly on small blocks of pixels, the noise reduction can be performed in real-time in a video processing pipeline.

There are disclosed an apparatus and method for processing images. The apparatus for processing images according to an embodiment includes an image input unit configured to receive a first image of a Bayer pattern including noise and an image output unit configured to output a noise-removed image by removing noise from the first image using a noise removal model, and the noise removal model includes a color correlation block configured to generate a second image of the Bayer pattern including RGB correlation information about the first image from the first image by performing rearrange and convolution operations on the first image, a discrete cosine transform (DCT) block configured to generate a third image in which a high-frequency component of the first image is highlighted by applying a DCT to the first image, and one or more discrete wavelet transform (DWT) blocks configured to generate a low-frequency feature map and one or more high-frequency feature maps by applying a DWT to a first feature map generated based on the first image, the second image, and the third image, and generate a final feature map in which a high-frequency component and a low-frequency component of the first feature map are highlighted based on the low-frequency feature map and the one or more high-frequency feature maps.

Metrology methods, modules and systems are provided, for using machine learning algorithms to improve the metrology accuracy and the overall process throughput. Methods comprise calculating training data concerning metrology metric(s) from initial metrology measurements, applying machine learning algorithm(s) to the calculated training data to derive an estimation model of the metrology metric(s), deriving measurement data from images of sites on received wafers, and using the estimation model to provide estimations of the metrology metric(s) with respect to the measurement data. While the training data may use two images per site, in operation a single image per site may suffice—reducing the measurement time to less than half the current measurement time. Moreover, confidence score(s) may be derived as an additional metrology and process control, and deep learning may be used to enhance the accuracy and/or speed of the metrology module.

A system for processing encoded image components for artificial intelligence tasks. The system can include one or more compute units, one or more controllers and memory. The one or more controllers can include one or more micro-op schedulers and one or more channel switches. The one or more compute units can be configured to process components of the transformed domain image data according to one or more micro-operations for an artificial intelligence task. The one or more channel switches can be configured to selectively control the transfer of the components of transformed domain image data to the one or more compute units based on one or more gating flags. The one or more channel switches can also be configured to selectively control generation of the one or more micro-operations by the one or more micro-op schedulers based on the one or more gating flags.

Methods and systems are provided for implementing static channel filtering operations upon image datasets transformed to frequency domain representations, including decoding images of an image dataset to generate a frequency domain representation of the image dataset; discarding coefficient values of one or more particular frequency channels of each image of the image dataset in a frequency domain representation; and transporting the image dataset in a frequency domain representation to one or more special-purpose processor(s). Methods and systems of the present disclosure may enable a filtered image dataset to be input to a second layer of a learning model, bypassing a first layer, or may enable a learning model to be designed with a reduced-size first layer. This may achieve benefits such as reducing computational overhead and time of machine learning training and inference computations, reducing volume of image data input into the learning model, and reducing convergence time.