公开(公告)号：WO2021074743A1

公开(公告)日：2021-04-22

申请号：PCT/IB2020/059379

申请日：2020-10-06

Applicant: SISVEL TECHNOLOGY S.R.L. ,  INSTITUT MINES TÉLÉCOM

Inventor： TARTAGLIONE, Enzo ,  GRANGETTO, Marco ,  ODIERNA, Francesco ,  BRAGAGNOLO, Andrea ,  FIANDROTTI, Attilio

IPC: G06N3/04 ,  G06N3/08

Abstract: The present invention relates to a method for pruning a neural network (100) comprising a plurality of neurons (105), said method comprising:- an initialization phase, wherein input information is fetched comprising at least parameters ({wni,bni}) related to said neural network (100) and a dataset (D) representative of a task that said neural network (100) has to deal with, wherein said parameters ({wni,bni}) comprising a weights vector (wni) and/or a bias (bni) related to at least one neuron (105) of said plurality of neurons;- a regularization phase, wherein said neural network (100) is trained according to a training algorithm by using said dataset (D);- a thresholding phase, wherein an element (wnij) of said weights vector (wni) is put at zero when its absolute value is below a given threshold (T),said method wherein, during said regularization phase, said parameters ({wni,bni}) evolve according to a regularized update rule based on a neural sensitivity measure (S) to drive towards zero parameters related to at least one less sensitive neuron (108) of said neural network (100), wherein said neural sensitivity measure (S) is based on a pre-activation signal of at least one neuron (105) of said plurality of neurons.

公开(公告)号：WO2018158738A1

公开(公告)日：2018-09-07

申请号：PCT/IB2018/051339

申请日：2018-03-02

Applicant: SISVEL TECHNOLOGY S.R.L.

Inventor： GRANGETTO, Marco ,  VERDOJA, Francesco

IPC: H03M7/30 ,  G06T9/20 ,  G06K9/48

Abstract: The present invention relates to a method for encoding the borders of pixel regions of an image, wherein the borders contain a sequence of vertices subdividing the image into regions of pixels (superpixels), by generating a sequence of symbols from an alphabet comprising the step of: - defining for each superpixel a first vertex for coding the borders of the superpixel according to a criterion common to all superpixels; - defining for each superpixel the same coding order of the border vertices, either clockwise or counter-clockwise; - defining the order for coding the superpixels on the base of a common rule depending on the relative positions of the first vertices; - defining a set of vertices as a known border ( Γ ), wherein the following steps are performed for selecting a symbol of the alphabet, for encoding the borders of the superpixels: a) determining the first vertex of the next superpixel border individuated by the common criterion; b) determining the next vertex to be encoded on the basis of the coding direction; c) selecting a first symbol ("0") for encoding the next vertex if the next vertex of a border pertains to the known border ( Γ ), d) selecting a symbol ("1"; "2") different from the first symbol ("0") if the next vertex is not in the known border (( Γ ); e) repeating steps b), c), d) and e) until all vertices of the superpixel border have been encoded; f) adding each vertex of the superpixel border that was not in the known border ( Γ ) to the set (( Γ ); g) determining the next superpixel whose border is to be encoded according to the common rule, if any; i) repeating steps a)-g) until the borders of all the superpixels of the image have being added to the known border ( Γ ).

公开(公告)号：WO2015019208A9

公开(公告)日：2015-02-12

申请号：PCT/IB2014/062727

申请日：2014-06-30

Applicant: SISVEL TECHNOLOGY S.R.L.

Inventor： PORZIO GIUSTO, Pietro

IPC: G06T3/00

Abstract: The present invention relates to an apparatus and a method for correcting the deformations that appear, in both bidimensional and stereoscopic vision, when images generated in compliance with the linear perspective rules are viewed from a point of view not corresponding to the center of projection of the perspective. The corrections are determined by a technique that tends to represent the images as if each point that constitutes them were captured with the lens aimed at it. In such hypothesis of lens aiming, the position gradient of the points to be represented is calculated, and the coordinates of the positions that the points must take in the image plane are obtained by integrating the components of such gradient. The application range of the method is very wide, due to the various embodiments and numerous variants it allows.

公开(公告)号：WO2015011612A1

公开(公告)日：2015-01-29

申请号：PCT/IB2014/063158

申请日：2014-07-16

Applicant: SISVEL TECHNOLOGY S.R.L.

Inventor： PASCHALAKIS, Stavros

IPC: G06K9/46 ,  G06T7/40

CPC classification number: G06K9/38 ,  G06K9/4642 ,  G06T9/00 ,  H04N19/14 ,  H04N19/30 ,  H04N19/40 ,  H04N19/90

Abstract: It is described a method for encoding an image descriptor (H) based on gradient histograms (hp ) into a transformed descriptor (V) comprising transformed subdescriptors (vp ), wherein each gradient histogram (hp) comprises a plurality of histogram bins ( hi ) and each subdescriptor (vp) comprises a set of values ( vj ), characterised in that said subdescriptors (vp ) are generated according to an element utilisation order list specifying a subdescriptor index (p)of said set of subdescriptors (vp) and an element index ( j ) of said set of values ( vj ).

Abstract translation: 描述了一种用于将基于梯度直方图（hp）的图像描述符（H）编码成包括变换的子描述符（vp）的变换描述符（V）的方法，其中每个梯度直方图（hp）包括多个直方图区段（hi） 并且每个子描述符（vp）包括一组值（vj），其特征在于，所述子描述符（vp）根据指定所述一组子描述符（vp）的子描述符索引（p）的元素利用顺序列表生成， 所述一组值（vj）的索引（j）。

公开(公告)号：WO2014203157A1

公开(公告)日：2014-12-24

申请号：PCT/IB2014/062286

申请日：2014-06-17

Applicant: SISVEL TECHNOLOGY S.R.L.

Inventor： SCOZZARO, Andrea ,  PREVITI, Gianluca

IPC: H04L29/06 ,  H04L12/58

CPC classification number: H04L63/1483 ,  H04L9/3247 ,  H04L51/12 ,  H04L63/0227 ,  H04L63/1441 ,  H04L63/168

Abstract: The present invention relates to a system (5) and a method for filtering electronic messages, wherein said system (5) comprises a client machine (1) that comprises a security module, wherein said security module has access to at least one message (4) sent by a sender identified in the message through respective identification data and received by said client machine (1), wherein said security module is configured for reading said sender's identification data from the header of said message (4), finding and storing any potentially harmful contents of the body of said message (4), verifying if said potentially harmful contents satisfy validation conditions which are specific for the sender having the identification data specified in the header of the message (4), allowing or denying access to at least part of the body of the message (4), in which said potentially harmful contents are present, based on the result of the verification.

Abstract translation: 本发明涉及一种用于过滤电子消息的系统（5）和方法，其中所述系统（5）包括客户机（1），其包括安全模块，其中所述安全模块可访问至少一个消息（4 ），由所述客户端机器（1）通过相应的标识数据由所述消息中标识的发送者发送，其中所述安全模块被配置为从所述消息（4）的头部读取所述发送者的标识数据，找到并存储任何潜在的 验证所述消息（4）的身体的有害内容，验证所述潜在有害内容是否满足对具有在消息（4）的头部中指定的标识数据的发送者特定的验证条件，允许或拒绝访问至少部分 的信息（4）的身体，其中存在所述潜在的有害内容，基于验证的结果。

公开(公告)号：WO2013128422A1

公开(公告)日：2013-09-06

申请号：PCT/IB2013/051658

申请日：2013-03-01

Applicant: SISVEL TECHNOLOGY S.R.L.

Inventor： STRIULI, Alessandro

IPC: H02J3/46

CPC classification number: H02J4/00 ,  G05B15/02 ,  H02J3/46

Abstract: A method is described for managing the supply of electric energy to a community of users (UP1, UP2,... UPN), said community having available at least one system for producing electric energy from a local source (100) and at least one system for supplying electric energy managed by an external distributor (110), comprising the steps of: determining a supply, to each user of said community of users (UP1, UP2,... UPN), of a combination of electric energy coming from said at least one system for producing energy from a local source (100) and from said at least one system for supplying energy managed by an external distributor (110), wherein said combination of electric energy can be determined specifically for each user; adjusting, for each user of said community of users (UP1, UP2,... UPN), the supply of the fraction of electric energy coming from said local source (100), as a function of the electric energy coming from said local source (100) to be distributed to the other users of said community of users (UP1, UP2,... UPN), so as to give priority to taking electric energy directly from said system for producing energy from a local source (100) over said at least one system for supplying electric energy managed by an external distributor (110). A device and a system are also described for managing the supply of electric energy to a user belonging to said community of users (UP1, UP2,... UPN).

Abstract translation: 描述了一种用于管理向用户群体（UP1，UP2，... UPN）供应电能的方法，所述社区具有至少一个用于从本地源（100）产生电能的系统和至少一个 用于提供由外部分配器（110）管理的电能的系统，包括以下步骤：向所述用户社区（UP1，UP2，... UPN）的每个用户确定来自 所述至少一个系统用于从本地源（100）和所述至少一个系统产生能量，用于供给由外部分配器（110）管理的能量，其中可以为每个用户专门确定所述电能组合; 对于所述用户社区（UP1，UP2，... UPN）的每个用户，调整来自所述本地源（100）的电能的分数，作为来自所述本地源的电能的函数 （100）分配给所述用户群体（UP1，UP2，... UPN）的其他用户，以便优先从所述系统直接从本地源（100）产生能量的系统 所述至少一个用于提供由外部分配器（110）管理的电能的系统。 还描述了一种用于管理向属于所述用户群体（UP1，UP2，... UPN）的用户提供电能的装置和系统。

公开(公告)号：WO2013054266A1

公开(公告)日：2013-04-18

申请号：PCT/IB2012/055475

申请日：2012-10-10

Applicant: CSP - INNOVAZIONE NELLE ICT SCARL ,  SISVEL TECHNOLOGY S.R.L.

Inventor： BORRI, Roberto ,  CRESPI, Floriana ,  PEROTTI, Alberto

IPC: H04L5/00 ,  H04L5/04 ,  H04L25/49

CPC classification number: H04L27/2601 ,  H04L5/0007 ,  H04L5/0041 ,  H04L5/0092 ,  H04L5/04 ,  H04L25/4902 ,  H04L27/2666

Abstract: The present invention relates to a method for generating a variable-band modulated signal (x'); the modulating digital signal (Χ') of said modulated signal (x') comprises at least one component (C, S, P); the digital information (R) about said variable band is repeatedly or periodically superimposed on said at least one component (C, S, P) or on said modulating signal (Χ'). While processing the received modulated signal (x"), the modulated digital signal (X") is extracted, the superimposed digital information (R) is obtained, and said extracted digital information (R) is used to determine the band of said modulated signal (x") and to complete the demodulation and decoding of said modulated signal (x").

Abstract translation: 本发明涉及一种用于产生可变频带调制信号（x'）的方法。 所述调制信号（x'）的调制数字信号（Chi'）包括至少一个分量（C，S，P）; 关于所述可变频带的数字信息（R）被重复地或周期地叠加在所述至少一个分量（C，S，P）上或所述调制信号（Chi'）上。 在处理接收到的调制信号（x“）的同时，提取调制数字信号（X”），获得叠加的数字信息（R），并且所述提取的数字信息（R）用于确定所述调制信号 （x“），并完成所述调制信号（x”）的解调和解码。

公开(公告)号：WO2013014646A1

公开(公告)日：2013-01-31

申请号：PCT/IB2012/053844

申请日：2012-07-27

Applicant: CSP – INNOVAZIONE NELLE ICT SCARL ,  SISVEL TECHNOLOGY S.R.L. ,  BORRI, Roberto ,  RICCHIUTI, Ferdinando ,  SCARAFIA, Simone ,  TORRERO, Livio

Inventor： BORRI, Roberto ,  RICCHIUTI, Ferdinando ,  SCARAFIA, Simone ,  TORRERO, Livio

IPC: G05D1/00

CPC classification number: G08G5/003 ,  G05D1/0022 ,  G08G5/0013 ,  G08G5/0026 ,  G08G5/0034 ,  G08G5/0039 ,  G08G5/0069 ,  H04L9/30

Abstract: A method is described for allowing missions of unmanned aerial vehicles (UAV), in particular in non-segregated air space, which comprises the steps of: prearranging a flight plan by an operator of an unmanned aerial vehicle, wherein the following steps are carried out: encrypting said flight plan, by said management and control body, with a private key of said management and control body, so as to obtain an encrypted flight plan; encoding said encrypted flight plan with a public key of the unmanned aerial vehicle for which said flight plan is intended, so as to obtain an encrypted and encoded flight plan.

Abstract translation: 描述了一种用于允许无人机（UAV）特别是在非隔离空域中的任务的方法，其包括以下步骤：由无人驾驶飞行器的操作者预先安排飞行计划，其中执行以下步骤 由所述管理机构用所述管理机构的私钥加密所述飞行计划，以获得加密的飞行计划; 用所述飞行计划所针对的无人驾驶飞行器的公钥对所述加密飞行计划进行编码，以获得加密和编码的飞行计划。

公开(公告)号：WO2012114320A1

公开(公告)日：2012-08-30

申请号：PCT/IB2012/050900

申请日：2012-02-27

Applicant: SISVEL TECHNOLOGY S.R.L. ,  PORZIO GIUSTO, Pietro

Inventor： PORZIO GIUSTO, Pietro

IPC: H04W64/00

CPC classification number: H04W4/023 ,  G01S11/02 ,  H04W64/00

Abstract: The present invention relates to a method for estimating the distance (d) of a receiver (102) from a radio transmitter (101), comprising the steps of: receiving (602) radio signals (103, 701) irradiated by the transmitter (101), which comprise components from which at least three tones (1,2,3,4) are extracted, each having a different frequency; measuring (606) a first phase difference (Δφ21) between first two tones (1, 2) of said at least three tones, whose frequencies (f1; f2) have a first spacing, and measuring a second phase difference (Δφ43) between second two tones (3, 4) of said at least three tones, whose frequencies (f3, f4) have a second spacing, wherein one of said first spacing or second spacing is greater than the other; estimating (607, 611, 613) said distance (d) on the basis of said first phase difference (Δφ21) and said second phase difference (Δφ43). The present invention also concerns related methods for calculating the position of a mobile terminal, to a mobile terminal, and to a device for localizing a mobile terminal. Said methods are preferably applied to transceiver systems using OFDM modulation, such as, for example, cellular networks compliant with the Long Term Evolution (LTE) standard and systems based on the IEEE 802.16 (WiMax) standard.

Abstract translation: 本发明涉及一种用于估计来自无线电发射机（101）的接收机（102）的距离（d）的方法，包括以下步骤：接收由发射机（101）照射的（602）无线电信号 ），其包括从其中提取至少三个音调（1,2,3,4）的组件，每个具有不同的频率; 测量（606）所述至少三个音调的前两个音调（1,2）之间的第一相位差（ωf21），其频率（f1; f2）具有第一间隔，并且测量第二相位差（Δf43） 在所述至少三个音调的第二两个音调（3,4）之间，其频率（f 3，f 4）具有第二间隔，其中所述第一间隔或第二间隔中的一个大于另一个; 基于所述第一相位差（Δf21）和所述第二相位差（Δf43）来估计（607,611,613）所述距离（d）。 本发明还涉及用于计算移动终端，移动终端的位置以及用于本地化移动终端的设备的相关方法。 所述方法优选地应用于使用OFDM调制的收发机系统，例如符合长期演进（LTE）标准的蜂窝网络和基于IEEE 802.16（WiMax）标准的系统）。

公开(公告)号：WO2012110938A1

公开(公告)日：2012-08-23

申请号：PCT/IB2012/050644

申请日：2012-02-13

Applicant: SISVEL TECHNOLOGY S.R.L. ,  DINI, Roberto

Inventor： DINI, Roberto

IPC: H04N5/44 ,  H04N7/26 ,  H04N13/00 ,  H04N21/83 ,  H04N21/44

CPC classification number: H04N13/106 ,  H04N13/139 ,  H04N13/156 ,  H04N13/178 ,  H04N13/189 ,  H04N19/46 ,  H04N19/597 ,  H04N21/235 ,  H04N21/435 ,  H04N21/44008 ,  H04N21/485 ,  H04N21/8126 ,  H04N21/816 ,  H04N21/83

Abstract: A method is described for acquiring and storing data relating to a video stream selectable by a user through a video processing apparatus (300), said method comprising the steps of: identifying the source of a video stream by means of at least one identification datum, said video stream having a coding and transport configuration adapted to produce a three-dimensional display; identifying, t he coding and transport configuration of said video stream through the acquisition of a configuration datum; establishing, by said video processing apparatus (300), an association between said at least one identification datum and said configuration datum and storing said association into storage means (12) of said video processing apparatus (300) for a subsequent selection of said, video stream.

Abstract translation: 描述了一种用于通过视频处理设备（300）获取和存储与用户可选择的视频流有关的数据的方法，所述方法包括以下步骤：借助于至少一个识别数据识别视频流的源， 所述视频流具有适于产生三维显示器的编码和传输配置; 通过获取配置数据来识别所述视频流的编码和传输配置; 由所述视频处理装置（300）建立所述至少一个识别数据和所述配置数据之间的关联，并将所述关联存储在所述视频处理装置（300）的存储装置（12）中，以便随后选择所述视频 流。