Abstract: The present disclosure relates to a method for encoding a video stream. The method comprising: acquiring pixel data of the video stream having a first resolution; extracting a crop corresponding to a region of interest from the pixel data of the video stream, the crop having the first resolution; down-scaling the pixel data of the video stream into a down-scaled video stream having a second resolution lower than the first resolution; processing the down-scaled video stream through one or more video processing operations; processing the crop through the one or more video processing operations; up-scaling the processed down-scaled video stream into an up-scaled video stream having the first resolution; merging the processed crop and the up-scaled video stream into a merged video stream; and encoding the merged video stream. The present disclosure further related to a video encoding device.

Abstract: A plurality of feature vectors for moveable objects are received in a sequence of image frames captured during a first period. The feature vectors are received from a machine learning module trained to extract similar feature vectors in different image frames. An initial value is assigned to a first feature vector , or to a cluster of feature vectors identified in a second sequence of image frames preceding the first image frame. The indicator indicates whether vector is alive. For each subsequent image frame , the indicator is updated by: updating the value based on a similarity between the feature vector and the first feature vector or the cluster of feature vectors. If the value of the indicator indicates that the vector is alive, determining that there is a moveable object that is located in the captured scene at least the predetermined portion of the given period of time.

Abstract: The present disclosure generally relates to a method for weighting of features in a feature vector of an object detected in a video stream capturing a scene, comprising: determining a feature vector comprising a set of features for a detected object in the video stream; acquiring a reference feature vector of a reference model of the scene; and assigning a weight to at least one feature of the determined feature vector, wherein the weight for a feature of the determined feature vector depends on a deviation measure indicative of a degree of deviation of the feature from a corresponding feature of the acquired reference feature vector of the reference model.

Abstract: A method for storing a file containing personal data comprises obtaining a temporary anonymous identifier (AnonID.m) of a person, which temporary anonymous identifier (AnonID.m) is dependent on a token (FileID.m) of the file; extracting personal data items from the file associated with the person; for each personal data item, generating a locator (Loc.m.n), which allows the personal data item to be reinstated into the file, and an item-specific anonymous identifier (AnonID.m.n) of the person, wherein the item-specific anonymous identifier is generated by applying a predefined one-way function to a combination of the temporary anonymous identifier (AnonID.m) and an identifier (n) of the personal data item; storing each personal data item together with the locator and the item-specific anonymous identifier in a first memory; and storing an anonymized version of the file without the personal data items in a second memory.

Abstract: A method for prioritization of video processing resources in a video processing and/or management system is provided. The method includes receiving video streams from multiple cameras and automatically ranking each camera as either a higher-priority or lower-priority camera. For higher-priority cameras, video content analysis is performed (as background processes), and the result is stored in storage. For lower-priority cameras, no such analysis is performed. The ranking is based on at least one of a historical usability, a historical cost, and a historical usability-to-cost ratio of a camera. A corresponding camera monitoring system, video processing and/or management system, computer program and computer program product are also provided.

Abstract: A method for encoding training data for training of a neural network comprises: obtaining training data including multiple datasets, each dataset comprises images annotated with at least one respective object class, forming , each dataset having an individual background class associated with the object class; encoding the images of the datasets to be associated with their respective individual background class; encoding image patches belonging to annotated object classes to be associated with their respective object class; encoding each of the datasets, to include an ignore attribute (“ignore”) to object classes that are annotated only in the other datasets and to background classes formed for the other datasets of the multiple datasets, the ignore attribute indicating that the assigned object class and background classes do not contribute in adapting the neural network in training using the respective dataset; and providing the encoded training data for training of a neural network.

Abstract: A method of generating an output video stream is provided, including receiving a first input video stream from at least one first camera capturing a scene, receiving a second input video stream from a second camera only partially capturing the scene, and generating an output video stream from the first input video stream, including, in response to determining that a particular area of the output video stream depicts a part of the scene depicted in the second input video stream, refraining from, or at least reducing a level of, a masking of the particular area of the output video stream. A corresponding device, camera monitoring system, computer program and computer program product are also provided.

Abstract: The present disclosure relates to a method performed by a deviation analyzing system (1) for black level drift assessment of a digital camera image sensor (21). The deviation analyzing system measures (1001) for pixels of an image captured by the image sensor, a respective luminance value and corresponding chromaticity value of each pixel. The deviation analyzing system further determines (1002) a respective average chromaticity value for differing luminance levels of the measured luminance values, based on for each luminance level averaging the corresponding measured chromaticity values. Moreover, the deviation analyzing system determines (1003), when the respective average chromaticity values of a range (4) of luminance levels indicate—to a predeterminable extent—chromaticity deviations (5) from the respective average chromaticity values (6) of other luminance levels (7), that a black level setting of the image sensor is drifted from a true black level.

Abstract: A method of encoding an image comprises establishing whether objects constituting one or more predefined object types or performing one or more predefined event types are visible in the image; in response to establishing that the objects are visible, encoding at least one region-of-interest of the image using a non-generative image model, thereby obtaining first image data; and encoding any remainder of the image using a generative image model, thereby obtaining second image data, wherein use of the non-generative image model enables decoding of the first image data without relying on information derived from images other than the encoded image or, if the image is a frame in a video sequence, enables decoding of the first image data without relying on information derived from images outside the video sequence.

Abstract: A method of activating an object-specific action comprises obtaining, based on image analysis of acquired images, a plurality of values of a movement characteristic, each value representing a moving object detected in the acquired images; receiving an input signal from a user input device, wherein the input signal represents a value of the movement characteristic for the user input device; comparing the value of the movement characteristic for the user input device with each value of the movement characteristics for the moving objects, and determining the moving object corresponding to the closest value of movement characteristic as a selected moving object; and activating the object-specific action for the selected moving object.

Abstract: A method in an access control system comprising a control unit connected to one or more mobile devices using radio-frequency signal communication. The control unit sends a motion status request to a connected mobile device out of the one or more connected mobile devices in response to the generated event. The method comprises determining an indication of a probability that the connected mobile device is associated with the event. Determining the indication is based on received signal strength data, measured by the connected mobile device on the radio-frequency signals transmitted by the control unit during a first predetermined period of time before the motion status request is received, and motion sensor data collected from a motion sensor comprised in the connected mobile device during a second predetermined period of time before the motion status request is received.

Abstract: A method for handling a radio-frequency connection between a control unit and a mobile device of an access control system. The mobile device is connected to the control unit using radio-frequency signal communication. The method comprises obtaining an indication of a first event generated in the access control system in response to a first request for the control unit to arrange to carry out an action. After a release of the connection by the control unit, the method further comprises obtaining an indication of whether or not a second event has been generated in the access control system in response to a second request for the control unit to arrange to carry out the action. The method further comprises determining, based on the indication of whether or not the second event has been generated, whether or not to allow the mobile device to re-connect to the control unit.

Abstract: A solid-state relay includes a semiconductor switch and a voltage boost block. The semiconductor switch has a control input, which causes the semiconductor switch to shift from an open, non-conducting position to a closed, conducting position when a voltage is applied to the control input. The voltage boost block includes a boost converter and a ground connector. A voltage output of the semiconductor switch is electrically connected to a voltage input of the boost converter. A voltage output of the boost converter is electrically connected to the control input. The ground connector of the boost converter is electrically connected to a voltage input of the semiconductor switch When the semiconductor switch is in the closed position, the semiconductor switch is maintained in a closed position in the absence of another control signal.

Abstract: There is provided a method for capturing a sequence of image frames in a thermal camera having a microbolometer detector comprising: capturing a first sequence and a second sequence of image frames with a shutter of the thermal camera being in a closed state and an open state, respectively. While capturing each of the first and the second sequence, an integration time of the microbolometer detector is switched between a plurality of integration times according to one or more repetitions of a temporal pattern of integration times. The method further comprises correcting image frames in the second sequence that are captured when the integration time is switched to a particular position within the temporal pattern of integration times using image frames in the first sequence that are captured when the integration time is switched to the same particular position within the temporal pattern of integration times.

Abstract: An image processor and a method therein to provide a target image for evaluation with an object detector. The method comprises: obtaining a source image captured by a camera and depicting an object, and applying an inverse pixel transform to each target pixel of a target image to determine one or more source pixels located at a position in the source image corresponding to a position of each target pixel in the target image. The method further comprises assigning, to each target pixel, a target pixel value determined based on one or more source pixel values of the determined one or more source pixels located at the corresponding position, thereby is a size c in the target image of the depicted object of a specific object type normalized in at least one size dimension. Thereafter, the target image is fed to an object detector for evaluation.

Abstract: A method of encoding a video stream for the provision of prunable video data, comprising: receiving, in an encoder, a video stream in which each image frame of the video stream is divided into a number of subviews, providing each subview with an assigned category based on a content of the particular subview, monitoring the evolution of category assigned for each subview, and encoding the video stream using a hierarchical P-frame encoding structure, in such manner that each change in category for a subview starts a new hierarchical P-frame structure.

Abstract: A method for encoding a video stream for the provision of prunable video data, comprising identifying, in the video stream, a first event-generating occurrence in one or more frames of the video stream, and, in an encoder, using the event-generating occurrence to initiate a hierarchical branch extending from a base-layer when encoding the video stream.

Abstract: A method of masking in an output image stream includes receiving an input image stream capturing a scene, processing the input image stream to generate the output image stream, including using a detector to detect objects in the scene and a tracker to track objects in the scene based on information provided by the detector, and further including to generate a particular output image of the output image stream by checking whether there exists a particular area in the scene in which an evaluation of a historical performance of the detector and/or tracker fulfills at least one condition, and to, if confirming that such a particular area exists, mask the particular area of the scene in the particular output image. A corresponding device, computer program, and computer program product are also provided.

Abstract: A method of masking or marking an object in an image stream is provided, including: generating one or more output image streams by processing an input image stream capturing a scene, including discarding pixel information about the scene provided by pixels of the input image stream, such that the discarded pixel information about the scene is not included in any output image stream; and detecting an object in the scene using the discarded pixel information, wherein generating the one or more output image streams includes masking or marking the detected object in at least one output image stream once deciding that the object is at least partially visible within this at least one output image stream. A corresponding device, computer program and computer program product are also provided.