公开(公告)号：US09992403B2

公开(公告)日：2018-06-05

申请号：US14319858

申请日：2014-06-30

Applicant: STMicroelectronics S.r.l.

Inventor： Salvatore Curti ,  Arcangelo Ranieri Bruna

IPC: G03B13/00 ,  H04N5/232 ,  G02B13/00

CPC classification number: H04N5/23212 ,  G02B13/009 ,  H04N5/23296

Abstract: In an embodiment, focusing an image-capture device such as, e.g., a camera including an optical system displaceable in opposite directions (A, B) via a focusing actuator, is controlled by evaluating a scale factor for the images acquired by the device. An accumulated value of the variations of the scale factor over a time interval (e.g., over a number of frames) is produced and the absolute value thereof is compared against a threshold. If the threshold is reached, which may be indicative of a zoom movement resulting in image de-focusing, a refocusing action is activated by displacing the optical system via the focusing actuator in the one or the other of the opposite focusing directions (A or B) as a function of whether the accumulated value exhibits an increase or a decrease (i.e., whether the accumulated value is positive or negative).

公开(公告)号：US20180082133A1

公开(公告)日：2018-03-22

申请号：US15472562

申请日：2017-03-29

Applicant: STMicroelectronics S.r.l.

Inventor： Nunziata Ivana Guarneri ,  Arcangelo Ranieri Bruna ,  Giuseppe Spampinato

IPC: G06K9/00 ,  G08G1/16 ,  G06T7/246 ,  G06T7/70 ,  G06T7/60

CPC classification number: G06K9/00791 ,  B60W30/095 ,  G06K9/00798 ,  G06K9/00805 ,  G06K9/00825 ,  G06T7/215 ,  G06T7/246 ,  G06T7/60 ,  G06T7/70 ,  G06T7/73 ,  G06T2207/30244 ,  G08G1/166 ,  G08G1/167

Abstract: A sequence of images is processed to generate optical flow data including a list of motion vectors. The motion vectors are grouped based on orientation into a first set of moving away motion vectors and a second set of moving towards motion vectors. A vanishing point is determined as a function of the first set of motion vectors and a center position of the images is determined. Pan and tilt information is computed from the distance difference between the vanishing point and the center position. Approaching objects are identified from the second set as a function of position, length and orientation, thereby identifying overtaking vehicles. Distances to the approaching objects are determined from object position, camera focal length, and pan and tilt information. A warning signal is issued as a function of the distances.

公开(公告)号：US09319586B2

公开(公告)日：2016-04-19

申请号：US14163090

申请日：2014-01-24

Applicant: STMicroelectronics S.r.l.

Inventor： Giuseppe Spampinato ,  Arcangelo Ranieri Bruna

IPC: H04N5/232 ,  G06T7/20

CPC classification number: H04N5/23254 ,  G06T7/246 ,  H04N5/23261 ,  H04N5/23267 ,  H04N5/23277

Abstract: According to an embodiment, a sequence of video frames as produced in a video-capture apparatus such as a video camera is stabilized against hand shaking or vibration by:—subjecting a pair of frames in the sequence to feature extraction and matching to produce a set of matched features;—subjecting the set of matched features to an outlier removal step; and—generating stabilized frames via motion-model estimation based on features resulting from outlier removal. Motion-model estimation is performed based on matched features having passed a zone-of-interest test confirmative that the matched features passing the test are distributed over a plurality of zones across the frames.

Abstract translation: 根据实施例，在诸如摄像机的视频捕获设备中产生的视频帧序列通过以下方式稳定，以防止手抖动或振动： - 在序列中插入一对帧以特征提取和匹配以产生一组 的匹配特征; - 将匹配的特征集合发布到异常值去除步骤; 并基于异常值去除产生的特征，通过运动模型估计产生稳定帧。 基于已经通过感兴趣区域测试的匹配特征来执行运动模型估计，证实通过测试的匹配特征分布在跨越帧的多个区域上。

公开(公告)号：US09100573B2

公开(公告)日：2015-08-04

申请号：US13841583

申请日：2013-03-15

Applicant: STMicroelectronics S.r.l.

Inventor： Giuseppe Spampinato ,  Arcangelo Ranieri Bruna ,  Giovanni Puglisi ,  Sebastiano Battiato

IPC: H04N5/228 ,  G06K9/40 ,  H04N5/232 ,  H04N5/14

CPC classification number: H04N5/23248 ,  H04N5/144 ,  H04N5/23254 ,  H04N5/23267

Abstract: A method and system for filtering an image frame of a video sequence from spurious motion, comprising the steps of dividing the image frame and a preceding image frame of the video sequence into blocks of pixels; determining motion vectors for the blocks of the image frame; determining inter-frame transformation parameters for the image frame based on the determined motion vectors; and generating a filtered image frame based on the determined inter-frame transformation parameters; wherein the image frame is dived into overlapping blocks.

Abstract translation: 一种用于从伪运动滤波视频序列的图像帧的方法和系统，包括以下步骤：将图像帧和视频序列的先前图像帧分割成像素块; 确定所述图像帧的块的运动矢量; 基于所确定的运动矢量确定所述图像帧的帧间变换参数; 以及基于所确定的帧间变换参数生成滤波图像帧; 其中图像帧被划分成重叠块。

公开(公告)号：US20150206324A1

公开(公告)日：2015-07-23

申请号：US14675630

申请日：2015-03-31

Applicant: STMICROELECTRONICS S.R.L.

Inventor： Angelo Bosco ,  Arcangelo Ranieri Bruna ,  Davide Giacalone ,  Rosetta Rizzo

IPC: G06T7/40 ,  G06T5/20 ,  G06T5/00

CPC classification number: G06T7/44 ,  G06T5/002 ,  G06T5/20 ,  G06T2207/10004 ,  G06T2207/10024 ,  G06T2207/20021 ,  G06T2207/20182

Abstract: An embodiment relates to a method for the detection of texture of a digital image, including providing a raw data image of the image by means of Bayer image sensors, determining noise in at least a region of the raw data image and determining the texture based on the determined noise without using a high pass or low pass filter.

Abstract translation: 一个实施例涉及一种用于检测数字图像的纹理的方法，包括通过拜耳图像传感器提供图像的原始数据图像，确定原始数据图像的至少一个区域中的噪声，并基于 确定的噪声，而不使用高通或低通滤波器。