公开(公告)号：US20240364859A1

公开(公告)日：2024-10-31

申请号：US18645593

申请日：2024-04-25

申请人： Siemens Healthineers AG

发明人： Alois Regensburger

IPC分类号： H04N13/398 ,  G06T3/40 ,  G06T5/20 ,  G06T5/70 ,  G06T5/73 ,  G06T7/50 ,  H04N13/383

CPC分类号： H04N13/398 ,  G06T3/40 ,  G06T5/20 ,  G06T5/70 ,  G06T5/73 ,  G06T7/50 ,  H04N13/383

摘要： For an improved stereopsis, a method and system is provided for stereo image processing and representation of one stereo image for a predetermined person. The stereo image includes (at least) one first partial image and one second partial image, which partial images represent different projection directions and are associated respectively with a right eye and a left eye of the person. The method includes: providing visual capacity data of the right eye and visual capacity data of the left eye and/or visual capacity data of both eyes of the predetermined person, processing the stereo image such that at least one corresponding image feature in the first partial image and in the second partial image is amended in the first partial image and the second partial image, respectively, dependent upon the visual capacity data of the respective associated eye of the predetermined person, and displaying the stereo image, processed in this manner, by a stereoscopic display apparatus.

公开(公告)号：US12131548B2

公开(公告)日：2024-10-29

申请号：US17769343

申请日：2020-04-15

申请人： ASELSAN ELEKTRONIK SANAYI VE TICARET ANONIM SIRKETI

发明人： Engin Uzun ,  Tolga Aksoy ,  Erdem Akagunduz

IPC分类号： G06V10/82 ,  G06T5/20 ,  G06V10/32 ,  G06V10/44 ,  G06V10/94 ,  G06V20/52 ,  G06V20/56

CPC分类号： G06V20/56 ,  G06T5/20 ,  G06V10/32 ,  G06V10/454 ,  G06V10/82 ,  G06V10/94 ,  G06V20/52 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06V2201/07

摘要： Disclosed is a method for training shallow convolutional neural networks for infrared target detection using a two-phase learning strategy that can converge to satisfactory detection performance, even with scale-invariance capability. In the first step, the aim is to ensure that only filters in the convolutional layer produce semantic features that serve the problem of target detection. L2-norm (Euclidian norm) is used as loss function for the stable training of semantic filters obtained from the convolutional layers. In the next step, only the decision layers are trained by transferring the weight values in the convolutional layers completely and freezing the learning rate. In this step, unlike the first, the L1-norm (mean-absolute-deviation) loss function is used.

公开(公告)号：US12131484B1

公开(公告)日：2024-10-29

申请号：US18732676

申请日：2024-06-04

申请人： Zhejiang University

发明人： Yunfeng Yan ,  Haoyuan Jin ,  Donglian Qi ,  Qi Li ,  Rui Han ,  Bingxiao Mei ,  Zezhou Wang ,  Gang Chen

IPC分类号： G06T7/20 ,  G06T5/20 ,  G06T7/11 ,  G06V10/44 ,  G06V10/80

CPC分类号： G06T7/20 ,  G06T5/20 ,  G06T7/11 ,  G06V10/44 ,  G06V10/806 ,  G06V2201/07

摘要： A multi-object tracking method based on authenticity hierarchizing and occlusion recovery relates to the technical field of multi-object tracking for security management and control of complex scenes. The method includes the following steps: image acquisition, object detection, authenticity hierarchizing, hierarchical association and outputting tracks. The multi-object tracking method is helpful to solve the long-term puzzling occlusion problem in the field of multi-object tracking, the constructed new algorithm framework can achieve an advanced tracking performance, improve the adaptability of the method to complex environments, and adapt to more difficult video surveillance security control tasks.

公开(公告)号：US20240354894A1

公开(公告)日：2024-10-24

申请号：US18437298

申请日：2024-02-09

申请人： Meta Platforms Technologies, LLC

发明人： Grant Kaijuin YANG ,  Steven Paul LANSEL ,  Irad RATMANSKY ,  Bruce ZITELLI

IPC分类号： G06T5/20 ,  G06F3/01 ,  G06T1/20

CPC分类号： G06T5/20 ,  G06F3/013 ,  G06T1/20

摘要： A filter system can effectively filter images with low computation cost by, instead of using large 2D filters, applying a series of smaller filters that require less compute and memory. In some implementations, the larger filter function, ƒ, is replaced by ƒ1, ƒ2, . . . , ƒN where ƒ(x)≈ƒ1(x)+ƒ2(x)+ . . . +ƒN(x). The filter system can apply these filters sequentially across multiple frames in time. The time integration of information by the human visual system results in the perception of a single higher quality filtering result, while using only the compute and memory footprint necessary to implement the filters. The number of frames across which a filter can be split without introducing flicker artifacts is dependent on the refresh rate of the display.

公开(公告)号：US12125180B2

公开(公告)日：2024-10-22

申请号：US17645028

申请日：2021-12-17

申请人： Lumenuity, LLC

发明人： Edo Waks ,  Benjamin Shapiro

IPC分类号： H04N23/50 ,  G02B5/10 ,  G02B17/08 ,  G06T5/20 ,  G06T5/80 ,  H04N23/80

CPC分类号： G06T5/80 ,  G02B5/10 ,  G02B17/0808 ,  G06T5/20 ,  H04N23/80

摘要： Described herein are systems and methods for reducing image aberrations in high magnification photography with partial reflectors. In particular, by an imaging device or camera that is built into or is included in a cell phone, smart phone, tablet, laptop or any other mobile device. The systems and methods include a light passing through a lens, a portion of said light then undergoes a number of partial reflections in-between two partial reflectors, and a portion of said light then reaches an imaging sensor. The partial reflections enable a longer light path to reach the imaging sensor, thus enabling a longer focal length to be used, which enables higher magnification. Described are methods and embodiments to select the physical parameters of optical elements in systems with partial reflectors, in order to create images with reduced image aberrations.

公开(公告)号：US12125174B2

公开(公告)日：2024-10-22

申请号：US17626162

申请日：2020-12-24

申请人： BOE TECHNOLOGY GROUP CO., LTD.

发明人： Guannan Chen ,  Ran Duan ,  Mengdi Sun ,  Lijie Zhang

IPC分类号： G06V10/77 ,  G06T3/4046 ,  G06T5/20 ,  G06T5/50 ,  G06T5/70 ,  G06V40/16

CPC分类号： G06T5/70 ,  G06T3/4046 ,  G06T5/20 ,  G06T5/50 ,  G06V10/7715 ,  G06V40/166 ,  G06V40/168 ,  G06T2207/20028 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/20221

摘要： An image processing method and device, and a computer-readable storage medium are disclosed. The method includes: performing a channel expansion process on the input image to obtain a first intermediate image; performing a channel decomposition process for multiple times based on the first intermediate image, wherein each time of channel decomposition process includes: decomposing an image to be processed into a first decomposition image and a second decomposition image; concatenating first decomposition images generated in each time of channel decomposition process and second decomposition image generated in the last time of channel decomposition process to obtain a concatenated image; performing a post-processing process on the concatenated image to obtain a second intermediate image; and fusing the second intermediate image with the input image to obtain the first output image.

公开(公告)号：US20240346649A1

公开(公告)日：2024-10-17

申请号：US18637087

申请日：2024-04-16

申请人： Boston Scientific Scimed Inc.

发明人： Erik Stephen Freed

IPC分类号： G06T7/00 ,  G06T5/20 ,  G06T5/70 ,  G06T5/90

CPC分类号： G06T7/0012 ,  G06T5/20 ,  G06T5/70 ,  G06T5/90 ,  G06T2207/10121 ,  G06T2207/20048 ,  G06T2207/30101

摘要： The disclosure provides methods and systems for medical imaging devices to route a path of a vessel on an extravascular image based on at least two points indicated on the image. An image speed map can be generated from the extravascular image and a list of the shortest path from points on the image to a selected point on the image can be identified from the image speed map. Then the path from another point on the image to the selected point on the image identified based on the identified shortest paths.

公开(公告)号：US20240331212A1

公开(公告)日：2024-10-03

申请号：US18584076

申请日：2024-02-22

申请人： CANON KABUSHIKI KAISHA

发明人： KOJI FUSE

IPC分类号： G06T11/00 ,  G06T5/20 ,  G06T7/50 ,  G06V10/141 ,  G06V10/44

CPC分类号： G06T11/00 ,  G06T5/20 ,  G06T7/50 ,  G06V10/141 ,  G06V10/44

摘要： A linear feature in an arbitrary direction on an object surface is detected. An image processing apparatus according to the present disclosure generates an image of a target object by lighting using a virtual light source and detects the linear feature included in the target object based on the image.

公开(公告)号：US20240331118A1

公开(公告)日：2024-10-03

申请号：US18378644

申请日：2023-10-10

申请人： AmTRAN TECHNOLOGY CO., LTD

发明人： Chao-Ching Chen

IPC分类号： G06T5/00 ,  G06T5/20 ,  G06T7/246 ,  G09G5/10

CPC分类号： G06T5/92 ,  G06T5/20 ,  G06T5/70 ,  G06T7/248 ,  G09G5/10 ,  G06T2207/20182 ,  G06T2207/20208 ,  G06T2207/20212 ,  G09G2320/0247 ,  G09G2320/0276 ,  G09G2320/103 ,  G09G2320/106 ,  G09G2370/20

摘要： A processing method for a plurality of image sources, for a display device includes inputting a plurality of image sources; detecting the plurality of image sources according to metadata of the a plurality of image sources to determine electro-optical transfer function (EOTF) corresponding to the a plurality of image sources; and dynamically applying different EOTFs on corresponding image sources.

公开(公告)号：US20240331110A1

公开(公告)日：2024-10-03

申请号：US18618317

申请日：2024-03-27

申请人： SparkCognition, Inc.

发明人： Elad Liebman ,  Alexandru Ardel ,  Ram Tuvi ,  Yash Gandhi ,  Dimitri Voytan

IPC分类号： G06T5/70 ,  G01V1/34 ,  G06T5/20

CPC分类号： G06T5/70 ,  G01V1/345 ,  G06T5/20

摘要： A method includes obtaining waveform return data including waveform return records for multiple sampling events associated with an observed area and generating image data based on the first subset of waveform return records. The method also includes reducing imaging artifacts in a region of interest of the image data using beam-domain local correction operations.