Abstract: An image convolution accelerator system including processing elements having kernel elements corresponding to a generated Toeplitz-like kernel. The Toeplitz-like generated kernel is based on a desired convolution kernel and padded with zeros, and an image controller is configured to activate the processing elements to multiply the kernel elements with image data when image data from a same row of the image is allocated to all processing elements corresponding to a non-zero row of the Toeplitz-like kernel. At least one storage element is configured to store at least the number of columns of the image minus a dimension of the desired convolution kernel+1.

Abstract: An image processing method including allocating a plurality of memory banks. Each memory bank includes a plurality of memory bank rows. The image is divided into a plurality of tags, wherein each tag has tag rows, each tag row has a plurality of pixels, and each tag row is consecutively distributed over different memory banks such that each tag row of the tag is accessible to be read out from the memory banks simultaneously. The consecutive distribution of the tag rows is performed by distributing the tag rows (r) in order starting from a first tag row (r) and to the memory banks (b) in order starting at the first memory bank (b), until all tag rows (r) have been distributed.

Abstract: A method (100) for creating a model for positioning, the method (100) comprising: receiving (S150) a training dataset (1) comprising samples (10) of an estimation labeled type, each sample (10) of the estimation labeled type being associated with a measurement position, each sample (10) of the estimation labeled type comprising sensor data (12), measured at the measurement position associated with the sample (10), the sensor data (12) being characteristic of the measurement position associated with the sample (10); an estimated measurement position (14), being an estimate of the measurement position associated with the sample (10); an estimated accuracy (16), being an estimate of an accuracy of the estimated measurement position (14); training (S160) a machine learning model (20) to convert sensor data (12) to a position, wherein the training is based on the samples (10) of the estimation labeled type in the training dataset (1), whereby the model for positioning is created.

Abstract: A stroke detection apparatus for determining that a stroke is occurring or has recently occurred to a user, wherein the apparatus comprises a data processing device comprising a processor and at least one wearable sensor configured to be worn by a user and configured to transmit a movement data of the user to the data processing device. The data processing device is configured to process the movement data to determine a first probability of a user stroke over a first time period and a second probability of a user stroke over a second time period. A stroke detection signal is then generated in dependence on at least the first probability and second probability.

Abstract: A method for generating an indoor environment model of a building comprises forming a transmitter location model for defining positions of transmitters in said building using gathered information for establishing transmitter locations (302), receiving signal strength indicative measurements being determined for a number of transmitters using at least one electronic communications device (304), wherein the signal strength indicative measurements are based on a signal which has varying signal propagation characteristics in the indoor environment, and wherein said signal strength indicative measurements are acquired from a number of known locations in the building, identifying discrepancies of signal transmittance in said indoor environment based on said signal strength indicative measurements in relation to said transmitter location model (306), determining locations of signal hindering elements causing said discrepancies (308), and generating said indoor environment model including transmitter locations and sa

Abstract: An apparatus being adapted to determine an occupancy in a designated area, said apparatus comprising a memory and a controller, said controller being configured to determine at least one passage time; wherein said passage time is a time for a detected passage into and/or out of said area based on said image stream, determine an average occupancy time, determine if a passage into said designated area is estimated to occupy the designated area by comparing the time for the passage with the average occupancy time, and determine an estimation of the present occupancy in the designated area to equal the number of such passages.

Abstract: An apparatus being adapted to determine an occupancy in a designated area, said apparatus comprising a memory and a controller, said controller being configured to determine at least one passage time; wherein said passage time is a time for a detected passage into and/or out of said area based on said image stream, determine an average occupancy time, determine if a passage into said designated area is estimated to occupy the designated area by comparing the time for the passage with the average occupancy time, and determine an estimation of the present occupancy in the designated area to equal the number of such passages.

Abstract: A method for tracking objects in a scene being viewed by a sequence of digital images comprises: separating a region of interest in the digital images into a plurality of spatially smaller, overlapping modules, and defining a hidden Markov model for each module. The method further comprises observing detections of positions of possible objects at different times, considering a set of states at a point of time (t1) such that the set of states comprises a state having a global optimum probability, and backtracking through the sequence of digital images for each hidden Markov model to shrink the set of states that are possible for earlier parts of the sequence of digital images, such that a single optimal state is found for an earlier point of time (t2<t1), whereby tracks of detected objects through the scene are determined up to the earlier point of time (t2).

Abstract: A method is presented for extracting information from a target area within a two-dimensional graphical object having a plurality of predetermined features with known characteristics in a first plane. An image is read where the object is located in a second plane, which is a priori unknown. A plurality of candidates to the features in the second plane are identified in the image. A transformation matrix for projective mapping between the second and first planes is calculated from the identified feature candidates. The target area of the object is transformed from the second plane into the first plane. Finally, the target area is processed so as to extract the information.