Video surveillance method and video surveillance system for carrying out said method

Abstract

The invention relates to a video surveillance method, referred to as the method. The method comprises a step of acquiring, using at least one imaging device, image streams of a space to be monitored, each image stream being acquired by the same imaging device. The acquisition step comprises acquiring at least one image stream, referred to as at least one dynamic image stream, by dynamically varying the focusing distance of the at least one imaging device, and / or acquiring, using at least two respective separate imaging devices, at least two image streams, referred to as at least two overlapping image streams, each of the at least two imaging devices being arranged in a different position. The method further comprises a step of selecting, from one or more images of at least one of the image streams, one or more areas of the space to be monitored. The method further comprises a step of obtaining images, referred to as sharp images, in which the sharpness of the one or more selected areas of the space to be monitored is improved, by modifying at least one of the images of the at least one of the image streams on the basis of at least one other of the images of the at least one of the image streams.

Description

DESCRIPTION Video surveillance method and video surveillance system for implementing such a method technical field

[0001] The present invention relates to the securing and monitoring of sites.

[0002] The technical field of the invention relates to a method and system for video surveillance. State of the art

[0003] State-of-the-art video surveillance systems consist of imaging a space to be monitored.

[0004] The acquired images are usually viewed in real time for monitoring by an agent assigned to monitor the site.

[0005] Images as acquired are also, in most cases, stored for a predetermined period of time.

[0006] When directly observing an imaged scene to be monitored, the security officer is, in most cases, limited to observing the images as acquired by the surveillance system.

[0007] No action allows it to quickly and satisfactorily verify whether suspicious movements or shapes in the images constitute a threat.

[0008] Indeed, at best, in some cases, the security guard has controls allowing them to modify the viewing angle, zoom, or focus distance of one camera at a time to concentrate their attention on a particular area. This action necessarily results in a change in the image field of view, that is, the area of ​​the site being monitored that is captured, and therefore the acquisition and observation of images (in real time) of a limited area of ​​the site; the entire area to be monitored is thus no longer captured.

[0009] Furthermore, in cases where the security officer can choose to change the area to be imaged, he must focus all his attention on the actions required to focus the camera, in the most appropriate way, on the desired area.

[0010] Furthermore, in the case of remotely operated surveillance systems, control of the surveillance system (choice of the area of ​​attention, the viewing angle, the zoom, or the focusing distance) is only possible by a single operator.

[0011] One object of the present invention is to remedy at least one of the drawbacks of the prior art.

[0012] Another objective of the invention is to provide a method and system for video surveillance: - allowing multiple users to obtain, independently of each other, a clear area of ​​their choice within the space to be monitored, and / or - allowing multiple users to simultaneously obtain a clear area of ​​their choice (and / or independently of each other) within the space to be monitored, and / or - allowing for multiple clear zones within the area to be monitored, and / or - allowing for a clear area within the monitored space while maintaining observation and / or acquisition of the entire monitored and / or imaged space, and / or - allowing all acquired images to be retained (in particular without reducing the field of view and / or the data available for subsequent processing), and / or - allowing the acquired and / or saved data (acquired and / or recorded images) to be enriched with additional data from the area to be monitored, and / or - allowing for enhanced images, particularly those with improved sharpness in certain areas, and / or - allowing for an increased field of view of the area to be monitored, and / or - allowing for increased depth of field in the area being monitored, and / or - allowing the display of the area to be monitored to be interactive. Description of the invention

[0013] For this purpose, a video surveillance method, called a process, is proposed.

[0014] The process includes a step of acquiring image streams, by at least one imaging device, of a space to be monitored.

[0015] Each image stream is acquired by the same imaging device.

[0016] The acquisition phase includes: • the acquisition of at least one image stream, referred to as at least one dynamic image stream, and / or • the acquisition of at least two image streams, referred to as at least two overlapping image streams.

[0017] Preferably, at least one dynamic image stream is acquired by dynamically varying the focusing distance of at least one imaging device.

[0018] Preferably, the at least two overlapping image streams are acquired by at least two separate respective imaging devices, each being arranged in a separate position.

[0019] The process includes a step of selecting one or more areas of space to be monitored.

[0020] Preferably, the selection step is implemented from one or more images from at least one of the image streams.

[0021] The process includes a step of obtaining images, called sharp images, in which the sharpness of the selected area(s) of the space to be monitored is improved.

[0022] The image acquisition step is implemented by modifying at least one of the images in at least one of the image streams based on at least one other image in at least one of the image streams.

[0023] It can be understood as "area to be monitored": a perimeter in the sense of a delimited or defined area of ​​space.

[0024] The term "selection of area(s) of space to be monitored" can be understood as one or more parts of the space to be monitored. Preferably, the term "selection of area(s) of space to be monitored" is understood as: ten, preferably nine, advantageously eight, particularly advantageously seven, particularly advantageously six, preferably five, even more preferably four, and even more preferably three areas of the space to be monitored. However, the invention may also include a selection of one, and / or two, areas to be monitored.

[0025] Preferably, the area(s) of space to be monitored are selected from one of the images in at least one of the image streams. The area(s) of space to be monitored may be selected from several of the images in at least one of the image streams.

[0026] Preferably, the area(s) of space to be monitored are selected from one of the image streams. The area(s) of space to be monitored can be selected from several image streams.

[0027] It can be understood as "image(s) from at least one of the image streams": one or more images from at least one dynamic image stream, and / or one or more images from at least two overlapping image streams.

[0028] The at least two overlay image streams may include one or more dynamic image streams.

[0029] Preferably, one, several, or each of the at least two overlay image streams can be a dynamic image stream.

[0030] Preferably, a "clear image" is understood to mean: an image obtained in which at least one or more parts of the space to be monitored have a sharpness greater than the sharpness of said part(s) of the image(s) acquired.

[0031] Preferably, a "clear image" is understood to mean: an image obtained in which at least one or more of the selected area(s) of the space to be monitored has a sharpness greater than the sharpness of said selected area(s) of the space to be monitored in at least one of the acquired image(s).

[0032] It can be understood as "depending on": relatively, based on, taking into account or in combination.

[0033] Preferably, the method includes displaying the sharp images on a display device.

[0034] Preferably, the method includes the display of one or more streams of sharp images.

[0035] Preferably, the display according to the method is carried out on one or more display devices.

[0036] Preferably, the method includes displaying the image streams acquired during the acquisition step on a display device.

[0037] Preferably, the method includes a display of one or more of the acquired image streams during the acquisition step.

[0038] Preferably, the acquisition step includes random dynamic modulation and / or sequential dynamic modulation of the focal distance of at least one imaging device from which at least one of the dynamic image streams is / are acquired.

[0039] It can be understood as "dynamic selection" or "dynamic modulation": a variation or change or dynamic modulation.

[0040] It can be understood as "sequential": according to a cycle, an order, a sequence or a given or predetermined pattern.

[0041] Preferably, the selection step includes a dynamic random selection and / or a dynamic sequential selection of the area(s) or each of the areas of the space to be monitored.

[0042] Preferably, the selection step includes a dynamic random selection and / or a dynamic sequential selection of one, areas or each of the predetermined areas of the space to be monitored.

[0043] Preferably, the selection step includes a dynamic random selection and / or a dynamic sequential selection of the area(s) of space to be monitored.

[0044] Preferably: the, or each of the predetermined zones are selected by a user, or the, the or each of the predetermined zones are selected, independently of each other, by several user(s), independently of each other.

[0045] Preferably: the area(s) or each of the areas of the space to be monitored are selected by one user, or the area(s) or each of the areas of the space to be monitored are selected, independently of each other, by several user(s), independently of each other.

[0046] Preferably, the process further includes a selection step, preferably based on the selected area(s), of at least one of the images from at least one of the image streams and / or at least one other of the images from at least one of the image streams.

[0047] The step of selecting at least one image from at least one image stream and / or at least one other image from at least one image stream can be carried out or implemented: - in a dynamically random and / or dynamically sequential manner, and / or - depending on the selected area(s), several of them, or each of them, and / or - based on depth information, for example a depth or disparity map, preferably of the selected area(s), and / or - depending on a focusing distance, preferably of the selected area(s), the area to be monitored, and / or - based on sharpness information, preferably of the selected area(s).

[0048] Preferably, at least one of the images from at least one of the image streams is selected from at least one dynamic image stream or from at least one overlay image stream and / or at least one other of the images from at least one of the image streams is selected from at least one dynamic image stream or from at least one overlay image stream.

[0049] Preferably, the process includes, for one, several or each of the images of one, several or each image stream, a comparison step.

[0050] In other words, the process may include a comparison step or a processing step including a comparison step.

[0051] Preferably, the comparison step is implemented for one, preferably several, preferably still for each of the images of one, preferably several, preferably still of each image stream.

[0052] Preferably, the comparison step includes or consists of comparing at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of one of the images of one of the image streams considered, preferably of one of the image streams among the at least two overlapping image streams or among the at least one dynamic stream, with at least one corresponding value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored, of another of the images, called the comparison image, of the reference stream or of another of the image streams, preferably of one of the image streams among the at least two overlapping image streams or among the at least one dynamic stream.

[0053] Preferably, according to a first alternative, the modification step, of at least one of the images of at least the image streams, includes an improvement in the sharpness of the image considered according to a deviation from, or measured or calculated or determined from, the comparison.

[0054] Preferably, each of the areas to be monitored consists of or corresponds, in each of the images comprising the, several or each of the areas to be monitored, to a group or block of pixels.

[0055] The process can be described as comprising a calculation or determination of the difference from the comparison.

[0056] Preferably, the modification step according to the first alternative, by improving the sharpness of the image considered according to a difference from the comparison, is called the image combination step (preferably the two-image combination step).

[0057] Preferably, the modification step, of at least one of the images of at least one of the image streams, preferably the image combination step, includes a minimization of a function.

[0058] Preferably, the function to be minimized includes, as a variable, the comparison of at least one value and / or, preferably, the difference resulting from the comparison.

[0059] Preferably, according to a second alternative, the modification of at least one of the images of at least one of the image streams, preferably of at least one dynamic image stream, includes, for a given image of one of the image streams considered, preferably of at least one dynamic image stream, an overlay in the given image, called the overview image, of said area(s) selected from, or extracted from, corresponding to or originating from, another image, called the identified image, of said image stream considered or of another of the image streams, preferably of at least one dynamic image stream.

[0060] Preferably, the selected area(s) in the identified image have a sharpness greater than or equal to the sharpness of the selected area(s) in the overall image.

[0061] Preferably, the overlay can be combined with the image combination step.

[0062] Preferably, one, several, or each of the at least two overlay streams can be a dynamic image stream.

[0063] The method may include a step of identifying an image, preferably at least one image, preferably the identified image, from at least one of the image streams, preferably from at least one dynamic image stream, in which the sharpness of the selected area(s) is greater than or equal to the sharpness of said area(s) selected in several other, preferably in each of the other, images from at least one of the image stacks, preferably from each of the image stacks.

[0064] The method may include a step of identifying an image, preferably at least one image, preferably the identified image, from at least one image stream, preferably from at least one dynamic image stream, by comparing the sharpness of the selected area(s) in a considered image, called the overall image, from at least one of the image stacks with the sharpness of the selected area(s) in several other images, preferably in each of the other images, from at least one of the image stacks, preferably from each of the image stacks.

[0065] According to the invention, a data processing device is also proposed comprising means arranged and / or programmed and / or configured to implement the process according to the invention.

[0066] According to the invention, a computer program is also proposed comprising instructions which, when the program is executed by a computer, lead the computer to implement the process according to the invention.

[0067] According to the invention, a computer-readable medium is also proposed comprising instructions which, when executed by a computer, lead the computer to implement the process according to the invention.

[0068] According to the invention, a video surveillance system, referred to as the system, is also proposed.

[0069] The system includes at least one imaging device arranged to image the space to be monitored.

[0070] Each of the imaging devices, among at least one imaging device, is arranged to acquire a stream of images of the space to be monitored.

[0071] The system includes a processing unit.

[0072] The processing unit is arranged and / or configured and / or programmed to: - to obtain, from one or more images from at least one of the image streams, one or more selected areas of the space to be monitored, and / or modify at least one of the images from at least one of the image streams, based on at least one other of the images from at least one of the image streams, to obtain sharp images in which the sharpness of the selected area(s) of the space to be monitored is improved.

[0073] Preferably, the system includes at least one arranged imaging device, or, in other words, at least one of the imaging devices of the system is arranged to acquire at least one dynamic image stream by dynamically varying the focusing distance of at least one imaging device.

[0074] Preferably, the processing unit is arranged and / or configured and / or programmed to modify at least one of the images in at least one of the image streams by overlaying it onto a considered image, referred to as the overview image, from one of the considered image streams of said selected area(s) in a another image, called identified image, from said image stream considered or from another of the image streams.

[0075] Preferably, the selected area(s) in the identified image have a sharpness greater than or equal to that of the selected area(s) in the overall image.

[0076] Preferably, the system includes at least two imaging devices arranged, or, in other words, at least two of the system's imaging devices are arranged, each in a distinct position and arranged to acquire a respective image stream, called an overlapping image stream.

[0077] Preferably, the processing unit is arranged and / or configured and / or programmed to, for one or more or each of the images of one, several or each image stream, compare at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of one of the images considered, called the reference image, of one of the image streams considered, called the reference stream, with at least one corresponding value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of another of the images of said image stream considered or of another of the image streams.

[0078] Preferably, the processing unit is arranged and / or configured and / or programmed to modify the image under consideration, based on a difference from the comparison, to improve the sharpness of the image under consideration.

[0079] The video surveillance system according to the invention may further include one or more display devices.

[0080] The video surveillance system according to the invention may be a drone.

[0081] The video surveillance system according to the invention can be a vehicle or a machine.

[0082] The processing device can be included in, or integrated into, any type of device, for example a user device, such as a smartphone, tablet, computer, calculator, processor, computer chip, programmed to implement the process according to the invention, for example by executing the computer program according to the invention.

[0083] Preferably, the data processing device and / or the user device and / or the VR / VA headset and / or the machine, vehicle and / or the drone according to the invention is suitable, preferably is particularly suitable, more preferably is designed and particularly advantageously is specially designed, to implement the method according to the invention.

[0084] Therefore, any characteristic of the process according to the invention is directly transposable to the data processing device and / or the user device and / or the VR / VA headset and / or the machine, vehicle and / or drone according to the invention and vice versa.

[0085] Preferably, the video surveillance system according to the invention is suitable, preferably even more so is particularly adapted, more preferably is designed, and most advantageously is specially designed, for implementing the video surveillance method according to the invention. Furthermore, any feature of the method according to the invention is directly transferable to the video surveillance system according to the invention, and vice versa.

[0086] Of course, the device according to the invention is not limited to the examples just given. Brief description of the FIGURES

[0087] The invention will be better understood upon reading the following description, given solely by way of non-limiting example and made with reference to the accompanying drawings in which: - Figure 1 is a schematic representation of a first embodiment of the video surveillance method according to the invention, - FIGURE 2 represents a flowchart illustrating the video surveillance process according to the first embodiment of the invention, - Figure 3 is a schematic representation of a second embodiment of the video surveillance method according to the invention, - FIGURE 4 represents a flowchart illustrating the video surveillance process according to the second embodiment of the invention, - FIGURE 5 is a schematic representation of an embodiment combining the first and second embodiments of the video surveillance method according to the invention, - FIGURE 6 represents a flowchart illustrating the video surveillance process according to the invention combining the first and second embodiments of the video surveillance process according to the invention.

[0088] It is understood that the embodiments described below are by no means exhaustive. In particular, variants of the invention may be conceived comprising only a selection of features described below, isolated from the other features described, if this selection of features is sufficient to confer a technical advantage or to differentiate the invention from the prior art. selection includes at least one functional preference feature without structural details, or with only a part of the structural details if that part is solely sufficient to confer a technical advantage or to differentiate the invention from the prior art.

[0089] In particular, all the variants and embodiments described can be combined with each other if there are no technical obstacles to this combination.

[0090] In the figures and in the rest of the description, elements common to several figures retain the same reference. Detailed description of the FIGURES

[0091] According to the non-limiting embodiment, an example of implementation of the video surveillance method according to the invention, referred to as the method, is described.

[0092] With reference to FIGURES 1 to 6, an embodiment of the process according to the invention is described.

[0093] The method includes a step of acquiring image streams, by at least one imaging device, of a space to be monitored. Each image stream is acquired by the same imaging device.

[0094] An imaging device is understood to be a camera or any device capable of acquiring a stream of images.

[0095] Each imaging device images or acquires its own stream of images.

[0096] The acquisition phase includes: - the acquisition of at least one image stream, referred to as at least one dynamic image stream, by dynamically varying the focal distance of at least one imaging device, and / or - the acquisition of at least two image streams, referred to as at least two overlapping image streams, by at least two separate imaging devices respective, each of the at least two imaging devices being arranged in a distinct position.

[0097] At least one dynamic image stream is acquired by dynamically varying the focal distance of at least one imaging device used or from which at least one dynamic image stream is acquired.

[0098] The at least two overlapping image streams are acquired by at least two separate imaging devices.

[0099] Each of the at least two imaging devices acquires or images a unique image stream from the at least two image streams.

[0100] Advantageously, each of the at least two imaging devices is arranged in a distinct position or location, or provides a different view of the space to be monitored.

[0101] The process also includes a selection step, called the area selection step, from one or more images of at least one of the image streams, of one or more areas of the space to be monitored.

[0102] Advantageously, the area selection step is implemented independently of the acquisition step.

[0103] The process also includes a step of obtaining images, called sharp images.

[0104] At least some of the sharp images obtained have improved sharpness, especially compared to the selected area(s) of the space to be monitored, preferably relative to the images acquired (or obtained during the acquisition stage).

[0105] In particular, the sharpness of the selected area(s) of the space to be monitored in the image(s) obtained is improved.

[0106] The sharp image(s) is / are obtained by modifying at least one of the images in at least one of the image streams according to at least one other of the images in at least one of the image streams.

[0107] Thus, and preferably, the area selection step(s) and / or the acquisition step and the step of obtaining clear images are implemented independently of each other.

[0108] In this way, each image stream can continue to be acquired independently of the selected area(s).

[0109] Also, unlike state-of-the-art devices and methods in which the observation of an area of ​​interest is carried out by voluntarily changing the focal distance (the zoom) of one or more camera(s) from displayed images to enlarge (or zoom in on or focus on) an area of ​​the space to be monitored, the invention makes it possible to reproduce clear images (or at least areas of clear images) in which selected areas (or areas of interest or areas of attention) appear clear without the recording of the image stream(s) by the camera(s) being modified or impacted (the recording of the image stream(s) continues independently of the choice of the area(s) to be reproduced clearly).

[0110] Advantageously, one, several, or each of the acquired image streams is displayed on one or more display devices. Typically, the display device is a screen. However, the display device could be a head-mounted display, for example, for virtual or augmented reality.

[0111] Advantageously, one, several or each of the sharp images are displayed on one or more display devices.

[0112] The acquisition step includes random dynamic modulation and / or sequential dynamic modulation of the focal distance of at least one imaging device from which at least one dynamic image stream is / are acquired.

[0113] In other words, one, several, or each of the imaging devices can be arranged to acquire images by dynamically varying the focusing distance.

[0114] Also, images can be acquired with a focal distance varying or being modulated according to a given or predetermined or predefined sequence or according to several sequences, succeeding each other for example in a redundant manner, given or predetermined or predefined.

[0115] Alternatively, or in combination, images can be acquired with a focal distance that varies or is modulated randomly. Thus, either all images are acquired with a randomly varying focal distance, or some images are acquired with a randomly varying focal distance and another set of images are acquired by varying or modulating the focal distance according to a given, predetermined, or predefined sequence.

[0116] Advantageously: the area(s) of the space to be monitored are selected by one user, or the area(s) of the space to be monitored are selected, independently of each other, by several user(s).

[0117] Advantageously, a user chooses, preferably in real time, the selected area(s).

[0118] Advantageously, each user chooses, preferably in real time, the area(s) they wish to monitor. This allows each user to select one or more areas of the space to be monitored without influencing the choices (or selections) made by other users, and without altering either the imaged area of ​​the space or the focal length used to image it.

[0119] It is also intended that the area(s) selected by one user may be shared with, or constitute one or more of, the selected areas of one, several, or all of the other users. Thus, a user noticing suspicious activity in one or more areas of the monitored space can, directly or indirectly, provide this information to the other users without contacting or alerting them directly and / or individually.

[0120] According to one embodiment, the area(s) of space to be monitored can be selected according to a random dynamic and / or a sequential dynamic.

[0121] In one embodiment, the area(s) of the space to be monitored are selected from among one or more predetermined or preselected areas of the space to be monitored. For example, the predetermined or preselected area(s) of the space to be monitored are sensitive areas of the space to be monitored (for example, building entrance areas, parking spaces, or storage areas).

[0122] According to this embodiment, the predetermined or preselected area(s) of the space to be monitored are selected according to a random dynamic and / or a sequential dynamic.

[0123] According to this embodiment, the predetermined or preselected area(s) of the space to be monitored are selected, for example in real time or episodically or when taking up duty by: a user, or independently of each other, by several user(s).

[0124] The process may include a selection step, called the image selection step, based on the selected area(s), of at least one of the images from at least one of the image streams and / or at least one other of the images from at least one of the image streams.

[0125] The image selection step can be implemented, for example, based on a value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored.

[0126] In this description, it can be understood as "value, data, or information contained in the pixels of the group or block of pixels": depth information, sharpness information, distance of focus, color data, texture data, brightness data, hue data, saturation data and / or RGB (for Red-Green-Blue) data.

[0127] Advantageously, image selection can be implemented based on depth information, sharpness information, and / or focal distance.

[0128] The method may include, for one, several or each of the images of one, several or each image stream, a step of comparing at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of one of the images considered of one of the image streams considered with at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of another of the images of said image stream considered or of another of the image streams.

[0129] Advantageously, the image selection step is implemented based on the comparison step and / or based on the difference resulting from the comparison.

[0130] In other words, the image selection step can be implemented based on or relative to at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored.

[0131] Advantageously, the comparison focuses on any information, value, or data contained within the pixels of the pixel matrix of the compared images. Preferably, but not limited to, this information may be color data, texture data, brightness data, hue data, saturation data, and / or RGB data.

[0132] According to an advantageous but non-limiting embodiment, the comparison step consists of detecting colors within a color matrix, by For example, a Bayer matrix is ​​used to find the best match between at least one image and the other image. In order to provide improved and / or precise depth information, it is advantageous to consider only the R (Red), VI (Green, or G (Green)), V2 (Green, or G (Green)), and B (Blue) pixels of the Bayer matrix corresponding to the reconstructed image, or to use pixels without a color filter with a particular geometry, called "dual pixels," to provide image depth sensitivity for the autofocus of the imaging system's lens.

[0133] However, the comparison step may consist of detecting one or more data contained in the pixels of the pixel matrix of the compared images from among a color data and / or texture data, and / or brightness data, and / or hue data, and / or saturation data, and / or RGB data.

[0134] According to a first non-limiting embodiment illustrated in FIGURES 1 and 2, the acquisition step includes the acquisition of a dynamic image stream ISl,t by dynamically varying the focal distance of a (single or unique) imaging device Cl.

[0135] As an example, the images of the dynamic image stream ISl,t are acquired at regular time intervals. The instant at which an image of an image stream is acquired is denoted t.

[0136] Advantageously, but not necessarily, two successive images (e.g. IS1,t1 and IS1,t2) within the same image stream have a distinct focal distance, denoted f.

[0137] According to the first non-limiting embodiment presented, the imaging device Cl has five focal distances, denoted fl, f2, f3, f4 and f5.

[0138] According to the first embodiment, the imaging device Cl can include two focusing distances f.

[0139] According to the first embodiment, it is also envisaged that the imaging device Cl has at least two focusing distances f.

[0140] According to the first embodiment, the modification of at least one of the images of at least one of the image streams includes an overlay in one of the images of at least one of the image streams, referred to as the overview image, of said selected area(s) in another image, referred to as the identified image, of said at least one of the image streams.

[0141] The selected area(s) in the identified image have a sharpness greater than or equal to the sharpness of the selected area(s) in the overall image.

[0142] Advantageously, but not necessarily, according to the first embodiment, the identified image is selected during the image selection step according to the invention. In other words, the identified image is selected based on the sharpness of the selected area(s).

[0143] Advantageously, the identified image is selected so that the selected area(s) of the space to be monitored have a sharpness greater than or equal to the sharpness of the corresponding selected area(s) in the overall image.

[0144] The overall image can be one of the images from one of the displayed image streams.

[0145] The identified image can be one of the images in the displayed image stream including the overall image or one of the images from another image stream.

[0146] Advantageously, the overall image and the identified image are acquired at a reduced time interval. In other words, according to the first embodiment shown in Figures 1 and 2, the overall image and the identified image are images belonging to the same image stream. Advantageously, the time interval separating the acquisition of the overall image from the acquisition of the identified image is less than a maximum time interval. The time interval is chosen so as to allow the same scene to be reproduced. In other words, in the event of movement of object(s) in the space being monitored, the maximum time interval separating the acquisition of the overall image and the acquisition of the identified image identified is sufficiently weak that moving objects do not exhibit significant positional change within the space being monitored.

[0147] Preferably, the maximum time interval is typically less than or equal to 5 seconds, preferably 4, preferably even more so 3, advantageously 2.5, particularly advantageously 2, most advantageously 1.75, preferably 1.5, even more preferably 1.25, more preferably 1, even more preferably 0.9, particularly preferred 0.8, most preferably 0.7 and most preferably less than or equal to 0.6 seconds.

[0148] According to a second non-limiting embodiment illustrated in FIGURES 3 and 4, the acquisition step includes the acquisition of three image streams IS1,t, IS2,t and IS3,t (from two or more image streams) by three respective distinct imaging devices (by two or more imaging devices) Cl, C2 and C3.

[0149] Advantageously, each of the imaging devices C1, C2, and C3 is positioned separately. Thus, each image stream is acquired from a different viewpoint of the space being monitored.

[0150] As an example, the images of the overlapping image streams IS1,t, IS2,t and IS3,t are acquired at regular time intervals. The instant at which an image of an image stream is acquired is denoted t.

[0151] According to the second non-limiting embodiment, each image in the same image stream has the same focal length f. According to the second illustrative embodiment, the images (for example, from each image stream) are acquired with the same focal length, denoted f3. Thus, each image in each image stream has the same focal length, denoted f.

[0152] However, the images in one of the image streams under consideration may have a focal length equal to or different from the focal length of the images in one, several, or each of the other image streams.

[0153] According to the second embodiment, the modification step includes an improvement in the sharpness of the image under consideration based on at least one of the images from at least one of the image streams and / or based on at least one other of the images from at least one of the image streams.

[0154] Advantageously, but not necessarily, the modified image of at least one of the image streams, according to the second embodiment, belongs to an image stream distinct from the image stream to which the other image belongs.

[0155] Advantageously, the modification step includes a combination (or reconstruction) of at least one of the images from at least one of the image streams based on, relative to, or from at least one other image (with another image according to the non-limiting embodiment) from the other image stream.

[0156] Advantageously, the modification step includes improving the sharpness of the image in question based on a discrepancy identified in the comparison step. Those skilled in the art are familiar with methods for improving image sharpness by comparing one image with one or more other images.

[0157] Advantageously, the modification step, according to the second embodiment, includes minimizing a function whose variable is the difference resulting from the comparison. Those skilled in the art are familiar with methods for improving image sharpness by minimizing the function comprising a comparison between two or more images.

[0158] The modification step can be defined as modifying the image by altering and / or substituting, in at least some of the pixels (especially when sharpness is insufficient) and / or moving a pixel or a set of pixels or one or more blocks of an image.

[0159] The comparison step consists of detecting any information, value, or data contained in the pixels of the pixel matrix of the images being compared.

[0160] As a non-limiting example, the modification step according to the second embodiment can be carried out by minimizing a function comprising: a term D, called the difference term, representing the difference or gap between the modified or combined or reconstructed image convolved by a point spreading function (PSF) and the image of the other image stream, and a term A, called the anomaly or penalty term, representing defects or anomalies within the combined (or reconstructed) image.

[0161] The term "penalties" guides the modification towards a preferred solution, for example, by preventing neighboring pixels from having values ​​that differ significantly from each other (except, for example, for neighboring block / contour pixels). This reduces the noise in the sharp image compared to at least one image in at least one image stream (which is potentially noisy), and also limits the effects of jagged edges on the contours of blocks or objects in the sharp image. The term "penalties" also improves sharpness information by preventing neighboring pixels from having excessively different values ​​and by removing outliers within the same block.

[0162] The difference term D can include the sum of several terms Di. Obtaining the distances Di obtained by comparing the reconstructed image convolved by the PSFs with at least one image from at least one of the image streams, can be achieved, among other ways, by evaluating the local colors in the pixels or groups of pixels opposite the discretization grid of the PSFs, so as to know the colors calculated at the positions of the photosites of the reconstructed image to make the distance comparisons at the appropriate places, without offset, allows maximum rendering accuracy.

[0163] The PSF can be substituted by any known function describing the response of an imaging system. For example, the PSF can be substituted by an optical transfer function (OTF) of the imaging system.

[0164] Preferably, the PSF is selected from a set of PSFs. Preferably, the PSF is selected based on one or more of the parameters on which the PSF depends.

[0165] Preferably, the function E, preferably again the term D and / or the term A, depends on the information of depth information and / or the sharpness information.

[0166] Advantageously, the modification step according to the second embodiment can include an iteration of the modification step according to the second embodiment.

[0167] In this case, the iterative minimization of the function having as its variable the difference from the comparison can be defined as including at least one iteration of the step of modifying at least one of the images of at least one of the image streams (in particular the reconstructed image) aimed at minimizing the function E.

[0168] Preferably, the iterative modification of the reconstructed image ends when the function E, or a combination of partial derivatives of the function E with respect to the reconstructed image being processed, is less than a minimization threshold, or when a certain number of iterations of the iterative modification of the reconstructed image is reached. The image thus reconstructed constitutes the sharp image.

[0169] According to a third, non-limiting embodiment illustrated in FIGURES 5 and 6, the acquisition step comprises the acquisition of three image streams (of two or more image streams) IS1,t, IS2,t and IS3,t by three respective separate imaging devices (by two or more imaging devices) Cl, C2 and C3.

[0170] Advantageously, each of the imaging devices C1, C2, and C3 is arranged in a distinct position. Thus, each of the image streams IS1,t, IS2,t, and IS3,t is acquired from a different viewpoint of the space being monitored.

[0171] As an example, images from image streams are acquired at regular time intervals. The instant at which an image from an image stream is acquired is denoted t.

[0172] According to the third non-limiting embodiment, advantageously, but not necessarily, two successive images (for example IS1,t1 and IS1,t2) within the same image stream have a distinct focal distance, denoted f.

[0173] According to the third non-limiting embodiment presented, the imaging devices Cl, C2 and C3 have five focusing distances, denoted fl, f2, f3, f4 and f5.

[0174] The third embodiment comprises or corresponds to a combination of the first and second embodiments.

[0175] Thus, any feature of the first embodiment, in particular each of the features of the first embodiment, is directly combinable with the second embodiment, in particular with each of the features of the first embodiment, and vice versa.

[0176] In other words, the modification step according to the invention can include obtaining an image by overlay, according to the first embodiment, then: improving the sharpness of the image obtained (by overlay) based on a difference resulting from the comparison between said image obtained and another image (from the image stream to which the image before overlay belongs or from another image stream), or improving the sharpness of another image from one of the image streams (belonging to the image stream to which the image before overlay belongs). or another of the image streams) based on a difference resulting from the comparison between said other image of said image stream and the image obtained (by overlay).

[0177] Similarly, the modification step according to the invention may include improving the sharpness of an image from one of the image streams based on a difference resulting from the comparison between said image from said image stream and another image from said image stream or another image stream, according to the second embodiment, and then: improving the sharpness of one of the images from one of the image streams (to which the image modified according to the second embodiment or another image stream belongs) by overlaying into said image from said image stream the selected area(s) from the image modified according to the second embodiment,or an improvement in the sharpness of the image modified according to the second embodiment by overlaying into said modified image (according to the second embodiment) the selected area(s) from one of the images of one of the image streams (to which the image modified according to the second embodiment belongs or from another of the image streams).

[0178] According to the invention, a data processing device is also proposed comprising means arranged and / or programmed and / or configured to implement the process according to the invention.

[0179] According to the invention, a computer program is also proposed comprising instructions which, when the program is executed by a computer, lead the computer to implement the process according to the invention.

[0180] According to the invention, a computer-readable medium is also proposed comprising instructions which, when executed by a computer, lead the computer to implement the process according to the invention.

[0181] According to the invention, a video surveillance system, referred to as the system, is also proposed.

[0182] The system includes at least one Cl, C2, C3 imaging device arranged to image the space to be monitored.

[0183] Each of the imaging devices Cl, C2, C3 is arranged to acquire an image stream IS1,t, IS2,t, IS3,t of the space to be monitored.

[0184] Preferably, but not necessarily, the system includes a processing unit.

[0185] The system can be autonomous.

[0186] The autonomous video surveillance system can be automated and / or embedded, meaning it is not necessarily operated and / or controlled by a user / operator / observer.

[0187] The monitoring system may include (and preferably not include a processing unit) means of communication arranged and / or programmed and / or configured to and / or capable of communicating with external means arranged and / or programmed and / or configured to implement all or part of the process according to the invention.

[0188] As an example, external means can be arranged and / or programmed and / or configured to implement the image acquisition step, and / or the step of selecting the area(s) of space to be monitored and / or the acquisition step and / or the comparison step and / or the step of selecting at least one of the images from at least one of the image streams and / or at least one other of the images from at least one of the image streams.

[0189] The video surveillance system processing unit can be arranged and / or programmed and / or configured to implement the image acquisition step, and / or the step of selecting the area(s) of space to be monitored and / or the acquisition step and / or the comparison step and / or the step of selecting at least one of the images from at least one of the image streams and / or at least one other of the images from at least one of the image streams, or to implement each of the steps of the video surveillance process according to the invention.

[0190] The video surveillance system may include a (single or unique) Cl imaging device.

[0191] In this case, the imaging device Cl is arranged to acquire a dynamic image stream ISl,t by dynamically varying the focal distance f.

[0192] Preferably, in this case, the video surveillance system is arranged to implement the process according to the first embodiment of the process according to the invention.

[0193] The video surveillance system may include three separate imaging devices (or two or more imaging devices) Cl, C2 and C3.

[0194] In this case, each imaging device is arranged to acquire a separate (overlap) image stream IS1,t, IS2,t and IS3,t.

[0195] Preferably, in this case, each of the images in the same image stream has the same focal length f.

[0196] Preferably, in this case, the video surveillance system is arranged to implement the process according to the second embodiment of the process according to the invention.

[0197] The video surveillance system may include three separate imaging devices (or two or more imaging devices) Cl, C2 and C3 and one, several or each imaging device Cl is arranged to acquire a dynamic image stream IS1,t, IS2,t, and / or IS3,t by dynamically varying the focusing distance f.

[0198] Preferably, in this case, the video surveillance system is arranged to implement the process according to the third embodiment of the process according to the invention.

[0199] Of course, the invention is not limited to the examples just described, and many modifications can be made to these examples without departing from the scope of the invention. Thus, in combinable variations of the embodiments described above: one, several, or each of the imaging devices can be arranged to be rotated, preferably relative to an axis of rotation or a pivot, so as to scan at least a part of the space to be monitored, and / or - preferably, when the imaging system comprises a single imaging system Cl, or according to the first embodiment, the single imaging device is arranged to scan at least a part of the space to be monitored, and / or one, several or each of the image streams can be stored, for example in a buffer, or RAM or cache memory, and / or in flash memory and / or in mass storage, for example a hard disk, and / or one, several or each of the images can be stored, for example in a buffer, or RAM or cache memory, and / or in flash memory and / or in mass storage, for example a hard disk.

[0200] In particular, all the variants and embodiments described can be combined with each other if there are no technical obstacles to this combination.

Claims

DEMANDS 1. A video surveillance method, referred to as the method, comprising: a step of acquiring image streams, by at least one imaging device, of a space to be monitored, each image stream being acquired by the same imaging device; said acquisition step comprises: • the acquisition of at least one image stream, referred to as at least one dynamic image stream, by dynamically varying the focal distance of at least one imaging device, and / or • the acquisition of at least two image streams, referred to as at least two overlapping image streams, by at least two respective separate imaging devices, each of the at least two imaging devices being arranged in a separate position, and a selection step, from one or more images of at least one of the image streams, of one or more areas of the space to be monitored, and a step of obtaining images, referred to as sharp images, in which the sharpness of the selected area(s) of the space to be monitored is improved, by modifying at least one of the images of at least one of the image streams according to at least one other of the images of at least one of the image streams.

2. A method according to claim 1, comprising displaying the sharp images on a display device.

3. Method according to claim 1 or 2, comprising displaying the image streams acquired during the acquisition step on a display device.

4. A method according to any one of the preceding claims, wherein the acquisition step comprises random dynamic modulation and / or sequential dynamic modulation of the focal distance of at least one imaging device from which at least one of the dynamic image streams is / are acquired.

5. A method according to any one of the preceding claims, wherein the selection step includes a random dynamic selection and / or a sequential dynamic selection of the area or areas of space to be monitored.

6. A method according to any one of claims 1 to 4, wherein the selection step includes a random dynamic selection and / or a sequential dynamic selection of one or more predetermined areas of the space to be monitored.

7. Method according to the preceding claim, wherein: the predetermined area(s) are selected by a user, or the predetermined area(s) are selected, independently of each other, by several user(s).

8. A method according to any one of claims 1 to 4, wherein: the area(s) of space to be monitored are selected by a user, or the area(s) of space to be monitored are selected, independently of each other, by several user(s).

9. A method according to any one of the preceding claims, further comprising a step of selecting, according to the selected area(s), at least one of the images from at least one of the image streams and / or at least one other of the images from at least one of the image streams.

10. A method according to any one of the preceding claims, comprising, for one, several or each of the images of one, several or each image stream, a step of comparing at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of one of the images considered of one of the image streams considered with at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of one of the other images of said image stream considered or of another of the image streams.

11. A method according to the preceding claim, wherein the modification step includes an improvement in the sharpness of the image considered as a function of a deviation from the comparison.

12. A method according to the preceding claim, wherein the modification step includes minimizing a function; said function comprising, as a variable, the difference resulting from the comparison.

13. A method according to any one of the preceding claims, wherein the modification of at least one of the images of at least one of the image streams comprises, for a given image of one of the image streams considered, an overlay in the given image, referred to as the overall image, of said or said selected area(s) in another image, referred to as the identified image, of said image stream considered or of another of the image streams; said or said selected area(s) in the identified image have a sharpness greater than or equal to the sharpness of said or said selected area(s) in the overall image.

14. Data processing device comprising means arranged and / or programmed and / or configured to implement the method according to any one of claims 1 to 13.

15. Computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method according to any one of claims 1 to 13.

16. Computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method according to any one of claims 1 to 13.

17. Video surveillance system, referred to as the system, comprising: - at least one imaging device arranged to image a space to be monitored; each imaging device being arranged to acquire an image stream of the space to be monitored, a processing unit arranged and / or configured and / or programmed to: • obtain, from one or more images from at least one of the image streams, one or more selected areas of the space to be monitored, and / or • modify at least one of the images in at least one of the image streams, based on at least one other image in at least one of the image streams, to obtain images, called sharp images, in which the sharpness of the selected area(s) of the space to be monitored is improved.

18. System according to the preceding claim, comprising at least one imaging device arranged to acquire an image stream, referred to as at least one dynamic image stream, by dynamically varying the focal length of at least one imaging device; the processing unit is arranged and / or configured and / or programmed to modify at least one of the images of at least one of the image streams by overlaying into a considered image of one of the considered image streams, referred to as the overall image, said selected area(s) in another image, referred to as the identified image, of said considered image stream or of another of the image streams; said selected area(s) in the identified image have a sharpness greater than or equal to said selected area(s) in the overall image.

19. System according to claim 17 or 18, comprising at least two imaging devices each disposed in a distinct position and arranged to acquire a respective image stream, referred to as the overlapping image stream; the processing unit is arranged and / or configured and / or programmed: - for one, several or each of the images of one, several or each image stream, compare at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of one of the images considered of one of the image streams considered with at least one value, data or information contained in the pixels of the group or block of pixels forming the area(s) to be monitored of another of the images of said image stream considered or of another of the image streams, and modify the image considered, according to a difference resulting from the comparison, to improve the sharpness of the image considered.

Images