公开(公告)号：US20210383501A1

公开(公告)日：2021-12-09

申请号：US17313221

申请日：2021-05-06

Applicant: Axis AB

Inventor： Viktor EDPALM ,  Song YUAN ,  Johan FÖRBERG ,  Viktors JAKOVELS

IPC: G06T3/00 ,  G06T7/55 ,  H04N5/225 ,  G06T3/40 ,  H04N5/232

Abstract: A digital camera and a method for aligning digital images comprising: receiving images including first and second images depicting a first and a second region of a scene, the regions being overlapping and displaced along a first direction; aligning the images using a transformation; determining disparity values for an overlap between the images; identifying misalignments by identifying blocks of pixels in the first image having a same position along a second direction and having disparity values exhibiting a variability lower than a first threshold and exhibiting an average higher than a second threshold; adjusting the transformation for the identified blocks of pixels in the first image and their matching blocks of pixels in the second image; and realigning the images using the adjusted transformation.

公开(公告)号：US20210358088A1

公开(公告)日：2021-11-18

申请号：US17242324

申请日：2021-04-28

Applicant: Axis AB

Inventor： Stefan LUNDBERG ,  Song YUAN ,  Tim BORGLUND

IPC: G06T5/00 ,  H04N5/232

Abstract: The present disclosure relates to a method performed by a background blurring system (1) for provision of a video compression stream from a video camera (2) adapted for capturing a scene. The background blurring system identifies (1001) in a first image of the scene, at a first point in time, at least a first pixel comprised in a background of the first image. The background blurring system further determines (1002) a blurred pixel value for the at least first pixel. Moreover, the background blurring system provides (1003), subsequent the first point in time, the blurred pixel value in the video compression stream. The background blurring system furthermore identifies (1004) in a second image of the scene, at a subsequent second point in time, that the at least first pixel has altered to be comprised in a foreground of the second image. Moreover, the background blurring system provides (1005) in the video compression stream, subsequent the second point in time—continuously and/or intermittently—identified non-blurred pixel values for the at least first pixel. The background blurring system further identifies (1006) in a third image of the scene, at a subsequent third point in time, that the at least first pixel has altered to be comprised in a background of the third image, wherein a value of the at least first pixel has remained unchanged from an intermediate point in time up to the third point in time. Furthermore, when the third point in time is within a predeterminable background merge time period from the intermediate point in time, the background blurring system provides (1007), subsequent the third point in time, the blurred pixel value in the video compression stream.
The disclosure also relates to a background blurring system in accordance with the foregoing, a video camera comprising such a background blurring system, and a respective corresponding computer program product and non-volatile computer readable storage medium.

公开(公告)号：US20210304367A1

公开(公告)日：2021-09-30

申请号：US17210555

申请日：2021-03-24

Applicant: Axis AB

Inventor： Song YUAN

IPC: G06T5/00 ,  G06T5/10 ,  G06T5/50 ,  H04N5/14 ,  H04N5/217

Abstract: The present disclosure relates to cameras and in particular methods for noise reduction in images captured by a camera, wherein a same perspective transform may be reused for performing temporal noise filtering on a plurality of images.

公开(公告)号：US20200327691A1

公开(公告)日：2020-10-15

申请号：US16839083

申请日：2020-04-03

Applicant: Axis AB

Inventor： Hampus LINSE ,  Song YUAN ,  Johan FÖRBERG

IPC: G06T7/73 ,  G06T5/50

Abstract: A method for detecting an object in a first distorted image using a sliding window algorithm, comprising: receiving an inverse of a mathematical representation of a distortion of the first distorted image; wherein the detection of an object comprises sliding a sliding window over the first distorted image, the sliding window comprising a feature detection pattern, and for each position of a plurality of positions in the first distorted image: transforming the sliding window based on the inverse of the mathematical representation of the distortion at the position, wherein the step of transforming the sliding window comprises transforming the feature detection pattern of the sliding window such that a resulting distortion of the feature detection pattern of the transformed sliding window corresponds to the distortion of the first distorted image at the position; and using the transformed sliding window comprising the transformed feature detection pattern in the sliding window algorithm.

公开(公告)号：US20240420466A1

公开(公告)日：2024-12-19

申请号：US18732991

申请日：2024-06-04

Applicant: Axis AB

Inventor： Song YUAN ,  Anton Öhrn ,  Niclas Danielsson

IPC: G06V20/40 ,  G06V10/776 ,  G06V10/82 ,  G06V20/52

Abstract: The present system and method generally relate to the field of camera surveillance, and in particular to object re-identification in video streams captured by a camera.

公开(公告)号：US20240221256A1

公开(公告)日：2024-07-04

申请号：US18532227

申请日：2023-12-07

Applicant: Axis AB

Inventor： Song YUAN ,  Ludvig Hassbring ,  Jessica Nilsson

IPC: G06T11/60 ,  G06T3/40 ,  G06T3/60 ,  H04N19/172 ,  H04N19/60

CPC classification number: G06T11/60 ,  G06T3/40 ,  G06T3/60 ,  H04N19/172 ,  H04N19/60

Abstract: A method for rendering a text string in a digital raster image suitable for block-wise transform coding, the method comprising: obtaining a partition of an image area into coding blocks for the block-wise transform coding; representing the text string as a plurality of graphical elements from a typeface which are arranged according to a tentative layout in the image area, wherein the tentative layout defines at least a position, orientation and size of each graphical element; modifying the tentative layout by applying a spectrum sparsening operation to at least one nonempty coding block, thereby obtaining a modified layout; and rendering a digital raster image of the graphical elements arranged according to the modified layout.

公开(公告)号：US20240205558A1

公开(公告)日：2024-06-20

申请号：US18537842

申请日：2023-12-13

Applicant: Axis AB

Inventor： Song YUAN ,  Johan FÖRBERG

IPC: H04N25/61 ,  G06T5/80 ,  G06V10/44

CPC classification number: H04N25/61 ,  G06T5/80 ,  G06V10/44

Abstract: A method for determining a lens distortion correction comprising:



obtaining an indication of a first feature and a first corrected position of the first feature in a first corrected image frame, the first corrected position is based on a first lens distortion correction of a first uncorrected position of the first feature in a first uncorrected image frame captured by the camera when equipped with a first interchangeable lens;
detecting a change of lenses on the camera;
obtaining an indication of a second feature and a second uncorrected position of the second feature in a second uncorrected image frame captured by the camera when equipped with a second lens;
matching the first feature with the second feature; and
determining a second lens distortion correction for the second lens based on a mapping between the first corrected position and the second uncorrected position.

公开(公告)号：US20240054598A1

公开(公告)日：2024-02-15

申请号：US18355784

申请日：2023-07-20

Applicant: Axis AB

Inventor： Song YUAN

IPC: G06T3/00 ,  G06T15/00 ,  G06T11/20

CPC classification number: G06T3/0018 ,  G06T15/00 ,  G06T11/203 ,  G06T2210/12 ,  G06T2200/24

Abstract: A method for defining an outline of a region in an image having distorted lines comprises displaying a first image version having distorted lines, wherein lines in a first image portion of the first version have a reduced distortion, receiving user input defining a first outline portion of the outline of region in the first version, wherein the first outline portion includes one or more lines in the first image portion of the first version, displaying a further version of the image having distorted lines, wherein lines in a further image portion of the further version have a reduced distortion, and receiving user input defining a further outline portion of the outline of the region in the further, wherein the further outline portion includes one or more lines in the further portion of the further version.

公开(公告)号：US20230410352A1

公开(公告)日：2023-12-21

申请号：US18333890

申请日：2023-06-13

Applicant: Axis AB

Inventor： Ludvig HASSBRING ,  Song YUAN

IPC: G06T7/70 ,  G06T3/40 ,  G06T7/20

CPC classification number: G06T7/70 ,  G06T3/40 ,  G06T7/20 ,  G06V2201/07 ,  G06T2207/20084 ,  G06T2207/20132

Abstract: A method and control unit for object detection in a video stream captured with an image acquisition device, especially for the purpose of anonymizing objects in the video stream.

公开(公告)号：US20230262238A1

公开(公告)日：2023-08-17

申请号：US18164069

申请日：2023-02-03

Applicant: Axis AB

Inventor： Viktor EDPALM ,  Song YUAN

IPC: H04N19/184 ,  H04N19/172 ,  H04N19/187 ,  H04N19/51 ,  H04N19/176

CPC classification number: H04N19/184 ,  H04N19/172 ,  H04N19/187 ,  H04N19/51 ,  H04N19/176

Abstract: A method of encoding a video stream including an overlay is provided, including: capturing a first image; adding an overlay to the first image at a first position, and encoding the first image in a first frame of a video stream; capturing a second image of the scene; determining a desired position of the overlay in the second image; encoding the second image in a second frame marked as a no-display frame, and generating and encoding a third frame including temporally predicted macroblocks at the desired position of the overlay referencing the first frame with motion vectors based on a difference between the desired position and the first position, and skip-macroblocks outside of the desired position of the overlay referencing the first frame. A corresponding device, computer program and computer program product are also provided.