PREDICTION OF ELECTRICAL CONSUMPTION INDUCED BY THE DISPLAY OF AN IMAGE BY A DIGITAL DISPLAY DEVICE

Abstract

A method for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device, said method comprising: obtaining a training base from a plurality of training images, each sample of said training base comprising: at least one characteristic belonging to a group of characteristics comprising: characteristics of a digital display device; characteristics of the environment of the digital display device; characteristics of a training image; a measured value of consumption induced by the display of said training image, said induced consumption value being measured on the digital display device during the display of the training image;training said supervised machine learning engine on said training base to minimize a loss function between the induced electrical consumption predictions made by the supervised machine learning engine and the measured induced consumption values ​​Abstract figure: Figure 1;

Description

Title of the invention: PREDICTION OF ELECTRICAL CONSUMPTION INDUCED BY THE DISPLAY OF AN IMAGE BY A DIGITAL DISPLAY DEVICE Technical field

[0001] The present disclosure relates to the field of digital image display. More specifically, it relates to the field of estimating the power consumption induced by the display of digital images. Prior art

[0002] Digital display devices are devices for broadcasting digital content. Digital content may, for example, include advertisements, but also information or awareness campaigns.

[0003] Digital display devices may, for example, be in the form of large screens placed in public spaces, for example in metro stations, on building fronts or integrated into travel shelters, in shop windows or in airports. Digital display devices may also be in the form of devices placed on the ground.

[0004] The display of images, or more generally of content by digital display devices, induces electrical consumption, in particular on the part of the screens displaying the images. With the deployment of numerous digital display devices in the urban environment, the total electrical consumption for the display of advertising campaigns can become significant. It is therefore desirable to evaluate and reduce the electrical consumption induced by the display of digital content on a set of digital display devices.

[0005] However, the amount of electrical energy consumed by digital display devices for the display of a given image or content may prove difficult to evaluate a priori. Indeed, it may vary depending on the content itself, the digital display device on which the content is broadcast, or even the environment of the digital display device.

[0006] Thus, the state of the art does not allow for the evaluation, and therefore reduction, of the electrical consumption induced by the broadcasting of content on digital screens.

[0007] There is therefore a need for a method for evaluating the electrical consumption induced by the broadcasting of content on digital screens. Summary

[0008] The present disclosure improves the situation.

[0009] A method for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device is proposed, said method comprising: obtaining a training base from a plurality of training images, each sample of said training base comprising: at least one characteristic belonging to a group of characteristics comprising: characteristics of a digital display device; characteristics of the environment of the digital display device; characteristics of a training image; a consumption value induced by the display of said training image, said induced power consumption value being obtained by means of a measurement carried out during the display of the training image;training said supervised machine learning engine on said training base to minimize a loss function between the induced electrical consumption predictions made by the supervised machine learning engine and the induced consumption values ​​obtained using the measurements made when displaying the images;

[0010] The term “electrical consumption induced by the display of at least one image by a digital display device” means the electrical consumption of the digital display device which is caused by the display of the image. This may be the electrical consumption of the digital display device when displaying the image, or the additional electrical consumption of the digital display device when displaying the image, compared to a basic electrical consumption of the device. The electrical consumption may be expressed in different ways. For example, it may be expressed in the form of an electrical power (for example in W) consumed when displaying the image, or an electrical energy (for example in joules) consumed for displaying the image.

[0011] “Training images” means images used, directly or indirectly, for training the supervised machine learning engine. These images can be used directly for training, when the supervised machine learning engine is trained to predict the electrical consumption at least from the images themselves, or indirectly, when they are used to measure the electrical consumption when reading the training images, but the supervised machine learning engine is trained to predict the consumption from characteristics not including the images themselves.

[0012] “Characteristic” means a characteristic relating to an image, a digital display device or the environment of a display device. digital used for the prediction of the power consumption induced by the display of the image.

[0013] The term “characteristic of a digital display device” means a characteristic of the device impacting the energy consumed when displaying an image. Such a characteristic may, for example, consist of: - a manufacturer of the device; - a model of the device; - a type of device screen (LCD, LED, etc.); - screen brightness; - a type of image processing used for displaying the image, for example a type of local dimming algorithm; - the age of the device; - etc.

[0014] “Characteristics of the environment of the digital display device” means physical characteristics of the immediate environment of the digital display device. A characteristic of the environment of the digital display device may, for example, consist of: - a position of the digital display device; - an orientation of the digital display device; - meteorological data, for example a temperature, humidity or ambient light value; - an ambient brightness value; - etc.

[0015] The term “characteristic of an image” means at least one piece of data characterizing the image. Such a characteristic may consist of: - digital pixel intensity values ​​of the image, for example: • the image itself; • light intensities of pixels on a layer representative of the image (for example a luminance layer); • a resized image; • etc. - global characteristics of the image, for example: • the image resolution; • the average brightness of the image; • an image contrast level; • etc. - etc.

[0016] The term “measured induced consumption value” means an induced consumption value resulting from a measurement of electrical consumption when displaying an image. This value may be the measured value itself (for example when the induced consumption is the electrical consumption of the digital display device when displaying the image), or a value obtained from the measured value (for example when the induced consumption is an additional electrical consumption of the digital display device when displaying the image, compared to a basic electrical consumption of the device).

[0017] The term “supervised machine learning engine” means a prediction function associated with parameters allowing: - during a so-called learning phase, to adapt the parameter values ​​in order to predict as accurately as possible the values ​​of annotated samples from a learning base; - during a so-called inference phase, to predict values ​​of the same type from unannotated samples.

[0018] According to the invention, different types of supervised learning engines can be implemented, such as for example: - one or more artificial neural networks; - support vector machines; - decision tree forests (or random forests, from the English "random forest”) ; - etc.

[0019] Thus, the supervised machine learning engine can be trained to be able to make accurate predictions of the power consumption induced by a given image on a given digital display device. This thus makes it possible to limit the power consumption induced by the display of images on digital display devices.

[0020] According to another aspect, there is provided a method for predicting the electrical consumption induced by the display of at least one image by a digital display device, said method comprising: obtaining at least one characteristic belonging to a group of characteristics comprising: characteristics of the digital display device; characteristics of the environment of the digital display device; characteristics of the at least one image to be displayed; executing a supervised machine learning engine taking as input said at least one characteristic, said supervised machine learning engine having been previously trained by a method according to one of the embodiments of the invention.

[0021] This makes it possible to predict, and therefore ultimately to reduce, the electrical consumption induced by the display of the image by the digital display device.

[0022] According to another aspect, there is provided a system for training a supervised machine learning engine for predicting the electrical consumption induced by the display of at least one image by a digital display device, said system comprising: at least one digital display device capable of measuring an electrical consumption induced by the display of at least one image; at least one calculation unit configured to execute a method according to one of the embodiments of the invention.

[0023] The term “computing unit” means an electronic component capable of performing computer calculations to perform a specific function. A computing unit may designate any type of processor or electronic component capable of performing digital calculations. For example, a computing unit may be an integrated circuit, an ASIC (from the English acronym “Application-Specific Integrated Circuit”, literally in French “integrated circuit specific to an application”, a microcontroller, a microprocessor, a DSP (from the English acronym “Digital Signal Processor”, literally in French “digital signal processor”), a processor, a GPU (from the English acronym “Graphics Processing Unit”, literally in French “graphics computing unit”). A computing unit according to the invention is not limited to a particular type of computing architecture. For example, a processor may implement a Harvard or Von Neumann type architecture.

[0024] The term "memory" means a digital electronic element used to store data. Different types of memories may be used in the invention, such as a read-only memory, a random access memory, a volatile memory or a flash memory. A device according to the invention may be equipped with one or more non-volatile memories which may be of different types such as mass memories, flash memory, read-only memories or SSD memories. A device according to the invention may also comprise one or more random access memories such as RAM, DRAM, SRAM, DPRREAM, VRAM, eDRAM or 1T-SRAM memories.

[0025] According to another aspect, there is provided a system for displaying at least one image, said system comprising: at least one digital display device; at least one computing unit configured to execute a method according to one of the embodiments of the invention.

[0026] According to another aspect, there is provided a digital display device deployed according to one of the embodiments of the invention.

[0027] According to another aspect, there is provided a supervised machine learning engine for predicting the electrical consumption induced by the display of at least one image by a digital display device driven by a method according to one of the embodiments of the invention.

[0028] According to another aspect, there is provided a computer program comprising instructions for implementing all or part of a method as defined herein when this program is executed by a processor.

[0029] According to another aspect, there is provided a non-transitory, computer-readable recording medium on which such a program is recorded.

[0030] The features set out in the following paragraphs may, optionally, be implemented, independently of one another or in combination with one another:

[0031] Advantageously, the at least one characteristic comprises the training image.

[0032] This allows the machine learning engine to be trained to predict the power consumption induced by the display of each image. A machine learning engine, in particular a neural network, will then be particularly capable of detecting the elements of an image (contrast, transitions, etc.) influencing the induced power consumption.

[0033] Advantageously, the at least one characteristic further comprises at least one characteristic of the digital display device on which the image is displayed.

[0034] This makes it possible to automatically learn the correlations between certain characteristics of the images, certain characteristics of the screens, and the consumption induced by the display of the images on the screens. This therefore makes it possible to improve the allocation of different images on different screens, in order to reduce the electrical consumption induced by the display of the images.

[0035] Advantageously, the at least one characteristic comprises a brightness value of a screen of the digital display on which the image is displayed at the time of displaying the image.

[0036] This makes it possible to predict the electrical consumption induced by the display of the image as a function of the brightness of the screen on which the image is displayed when the image is displayed.

[0037] Advantageously, the at least one characteristic comprises at least one characteristic of the environment of the digital display device.

[0038] This makes it possible to predict the consumption induced by the display of an image depending on the arrangement of the device in its environment.

[0039] Advantageously, the at least one characteristic of the environment of the digital display device comprises at least one meteorological datum.

[0040] This makes it possible to predict the consumption induced by the display of an image based on the meteorological characteristics observed or predicted at a given time.

[0041] Advantageously, the at least one characteristic comprises at least one characteristic of the digital display device or of the environment of the device; the method comprises a modification of said at least one characteristic of the digital display device or of the environment of the device to reduce the prediction of the electrical consumption induced by the display of the at least one image.

[0042] This makes it possible to modify a digital display device or its environment in order to reduce the electricity consumption induced by the display of images on the digital display device.

[0043] Advantageously, the method further comprises deploying the device according to said at least one modified characteristic.

[0044] “Deployment of the digital display device” means the installation of a device in an environment according to given characteristics. For example, the characteristics may be characteristics of the device itself (size, type of screen, etc.) or characteristics of its environment (position, orientation, etc.).

[0045] This enables the digital display device to be deployed according to the modified characteristics, so that the display of images on the device as deployed is carried out with reduced power consumption.

[0046] Advantageously, the at least one characteristic comprises at least one characteristic of the at least one image to be displayed; the method further comprises modifying said at least one characteristic of the at least one image to be displayed to reduce the prediction of the electrical consumption induced by the display of the at least one image.

[0047] This makes it possible to modify at least one image, so as to reduce the electrical consumption induced by the display of the image.

[0048] Advantageously, the method further comprises displaying the at least one modified image on a digital display device.

[0049] By "displaying the at least one modified image on a digital display device" is meant an action causing the at least one image to be displayed on a digital display device. For example, this may involve the physical display of the image itself, or sending a display instruction to the device, for example sending the at least one image itself accompanied by a display instruction.

[0050] This makes it possible to display a modified image, so as to reduce the electrical consumption induced by the display of the image.

[0051] Advantageously, the method comprises: predicting the electrical consumption induced by the display of a plurality of contents comprising at least one image by at least one digital display device; selecting at least one content to be displayed from among said plurality of contents, in order to minimize the prediction of the electrical consumption induced by the display of the at least one content.

[0052] This makes it possible to select content from among several possible contents, which minimize the consumption induced by its display.

[0053] Advantageously, the method further comprises displaying the at least one selected content on the at least one digital display device.

[0054] This makes it possible to display selected content so that its display induces reduced power consumption. Brief description of the drawings

[0055] Other characteristics, details and advantages will appear on reading the detailed description below, and on analyzing the attached drawings, in which: Fig.l

[0056] [Fig.l] shows a system for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device according to a set of embodiments of the invention. Fig. 2

[0057] [Fig.2] shows a method of training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device according to one embodiment. Fig. 3

[0058] [Fig.3] shows a system for displaying at least one image by at least one digital display device according to a set of embodiments of the invention. Fig. 4

[0059] [Fig.4] shows a method for predicting the electrical consumption induced by the display of at least one image by a digital display device according to one embodiment. Fig. 5a

[0060] [Fig.5a] shows an example of electrical consumption induced by the display of images of a first content, as measured, according to one embodiment. Fig. 5b

[0061] [Fig.5b] shows an example of electrical consumption induced by the display of images of a first content, as predicted, according to one embodiment. Fig. 6a

[0062] [Fig.6a] shows an example of power consumption induced by the display of images of a second content, as measured, according to one embodiment. Fig. 6b

[0063] [Fig.6b] shows an example of power consumption induced by the display of images of a second content, as predicted, according to one embodiment. Fig. 7

[0064] [Fig.7] shows an example of a method for predicting electricity consumption induced by the display of images on a digital display device of a device based on at least one characteristic of the device or its environment according to one embodiment. Fig. 8

[0065] [Fig.8] shows an example of a method for modifying at least one characteristic of at least one image to reduce the electrical consumption induced by the display of the at least one image, according to a set of embodiments of the invention. Fig. 9

[0066] [Fig.9] shows an example of a method of selecting content to be displayed according to an embodiment. Fig. 10

[0067] [Fig. 10] shows an example of a method of selecting a device on which to display content according to one embodiment Description of the embodiments

[0068] Reference is now made to [Fig. 1].

[0069] [Fig.l] shows a Sysl system for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device according to a set of embodiments of the invention.

[0070] The Sysl system comprises at least one digital display device capable of measuring an electrical consumption induced by the display of at least one image during the display of the image.

[0071] In the example of [Fig.l], the system Sysl comprises three digital display devices Displ 1, Displ2 and Displ3, each equipped respectively with a screen Ecrl 1, Ecrl2, Ecr 13. This example is provided as a non-limiting example only, and a variable number of digital display devices may be implemented. The devices may be of the same type or of different types (for example, may correspond to a single model of digital display device, or to different models).

[0072] The digital display devices Displ 1, Displ 2 and Displ 3 are capable of displaying at least one image, or more generally broadcasting content comprising one or more images, and of measuring the electrical consumption during the display of the image. This thus makes it possible to determine, for a given image and digital device, an electrical consumption value induced by the display of the image on the digital display device.

[0073] In the example of [Fig.l], the digital display devices are fixed devices intended to broadcast advertising content. The invention is however not limited to this example. For example, the digital display devices can be any type of digital device capable of displaying images. For example, they can be portable devices such as smartphones or tablets.

[0074] The Sysl system further comprises at least one computing device, for example at least one Servi server. In the example of [Fig.l], a single Servi server is shown. However, this example is provided as a non-limiting example only, and the invention is not restricted to a particular type of computing device. According to different embodiments of the invention, the computing and memory units may be distributed over a single computing device, or over several computing devices, for example several servers. The computing devices may also be of different types. For example, training may be carried out on a personal computer rather than on a server.

[0075] The at least one computing device, for example the server Servi, may be in communication with the digital display devices via a Connl connection. The Connl connection may comprise any type of link enabling data to be exchanged between the at least one computing device and the digital display devices. For example, the Connl connection may comprise a wired (fiber optic, ADSL, etc.) or wireless (Wi-Fi, 4G, 5G, etc.) connection, thus enabling the at least one computing device to send images to be displayed to each of the digital display devices, and to receive a power consumption value during the display of each image.If the consumption value induced by the display of said training image is not the gross consumption value of the display device during the display of the image, the calculation of the induced consumption value can be done, either by the display device itself, or by the at least one calculation device, from the gross electricity consumption value.

[0076] The Sysl system comprises at least one memory capable of storing: - a Motl supervised machine learning engine; - a Bas 1 training base.

[0077] As indicated above, a supervised machine learning engine is a prediction function, or model, associated with parameters. For example, the supervised learning engine may comprise an artificial neural network associated with parameters such as the weights of the connections between neurons or activation thresholds of sigmoid functions. The storage of the supervised machine learning engine Motl therefore corresponds to the storage of the prediction function and the parameters.

[0078] The training base Basl comprises all the data used to train the model. The training base thus comprises a set of training samples, each sample corresponding to the display of a given training image on a given digital display device. Each sample thus comprises: - at least one characteristic belonging to a group of characteristics comprising: • characteristics of a digital display device; • characteristics of the environment of the digital display device; • characteristics of the training image; - a measured value of consumption induced by the display of said training image, said induced electrical consumption value obtained, directly or indirectly, by a measurement carried out on the digital display device during the display of the training image.

[0079] The at least one computing device also comprises at least one computing unit Calcl configured to train the supervised machine learning engine, for example by implementing a method according to one of the embodiments described in [Fig.2],

[0080] [Fig.2] shows a method of training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device according to one embodiment.

[0081] The method P2 can for example be implemented by a calculation unit such as the calculation unit Calcl.

[0082] The method P2 comprises a first step S21 of obtaining a training base from a plurality of training images. As indicated above, each sample of the training base is associated with a training image and a digital display device, and comprises: - at least one characteristic; - a consumption value induced by the display of the image on the display device obtained by means of a measurement carried out on the digital display device when displaying the training image.

[0083] Obtaining the training base can therefore be done by loading a pre-constituted base, and / or creating all or part of the base.

[0084] The creation of the base can be done, for each sample: - on the one hand, by extracting at least one characteristic; - on the other hand, by displaying the training image on the digital display device, and by measuring the consumption during the display, in order to obtain, directly or indirectly, the electrical consumption induced by the display of the training image.

[0085] The induced electricity consumption can, according to different embodiments, be carried out on one or more electricity consumption measurements relating to: - the electricity consumption of the entire display device; - the power consumption of the screen; - the power consumption of other display device elements than the screen used to display the image: router, computing units, video players, fans, etc.

[0086] Thus, the induced consumption may correspond, according to different embodiments of the invention, to the consumption of the screen alone, or to the consumption of a set of elements of the display device involved in reading the training image. The induced electrical consumption value retained may be the raw electrical consumption value measured (or the sum of the values ​​measured on several elements are taken into account), or a value deduced from the measured values. For example, the induced consumption may be obtained by subtracting from the total measured consumption of the device when displaying the image a value representative of a basic electrical consumption of the display device.

[0087] The method P2 then comprises a second step S22 of training the supervised automatic learning engine on said training base to minimize a loss function between the induced electrical consumption predictions made by the supervised automatic learning engine and the induced consumption values ​​obtained using the measurements made when displaying the images.

[0088] In practice, this step consists of adapting the parameter values ​​of the supervised machine learning engine, in order to minimize a loss function representing the difference between the predictions of the learning engine and the induced consumptions resulting from the measurements. For example, If the supervised machine learning engine is an artificial neural network, the second step S22 can be performed through a technique called gradient backpropagation.

[0089] Thus, the supervised machine learning engine can be trained to be able to make accurate predictions of the power consumption induced by a given image on a given digital display device. This thus makes it possible to limit the power consumption induced by the display of images on digital display devices.

[0090] According to different embodiments of the invention, different characteristics, or combination of characteristics, can be taken into account.

[0091] Generally, different features can be selected depending on the desired prediction.

[0092] For example, the at least one feature may include the training image itself.

[0093] This allows the machine learning engine to be trained to predict the power consumption induced by the display of each image. A machine learning engine, in particular a neural network, will then be particularly capable of detecting the elements of an image (contrast, transitions, etc.) influencing the induced power consumption.

[0094] The at least one characteristic may also comprise, rather than the image itself, one or more characteristics representative of the image (resolution, dimension, contrast, etc.) which can serve as predictors of the electrical consumption induced by the display of the image.

[0095] In one set of embodiments of the invention, the at least one characteristic further comprises at least one characteristic of the digital display device on which the image is displayed.

[0096] In other words, the characteristics serving as bases for the prediction include in this case the image itself (or a distinctive characteristic of the image), and at least one characteristic of the display device. For example, certain image characteristics may be associated with higher consumption for certain types of screens, certain manufacturers, certain screen brightness values, certain types of local dimming algorithms, etc.

[0097] The machine learning engine will thus be able to make the link between the characteristics of the images and the screens, in order to correlate certain combinations of characteristics with a higher or lower induced consumption. This therefore makes it possible to improve the allocation of the different images to different screens, depending on the characteristics of the images and the screens, in order to optimize the allocation of the images to display on different screens to reduce the power consumption induced by the display of images.

[0098] This makes it possible to automatically learn the correlations between certain characteristics of the images, certain characteristics of the screens, and the consumption induced by the display of the images on the screens. This therefore makes it possible to improve the allocation of different images on different screens, in order to reduce the electrical consumption induced by the display of the images.

[0099] The characteristics of the digital display device on which the image is displayed may include, in a non-limiting manner, one or more characteristics chosen from: - a manufacturer of the device; - a model of the device; - a type of device screen (LCD, LED, etc.); - screen brightness; - a type of image processing used for displaying the image, for example a type of local dimming algorithm; - age of the device; - etc.

[0100] The at least one characteristic may also include a brightness value of a screen of the digital display on which the image is displayed at the time the image is displayed.

[0101] Indeed, the electrical consumption induced by the display of the image can also depend on a brightness setting of the screen when the image is displayed. This characteristic therefore makes it possible to predict the consumption induced by the display of the image as a function of the brightness parameter used.

[0102] The machine learning engine may learn to make induced consumption predictions based on the screen brightness value alone, or in addition to other characteristics such as the characteristics of the image or the display device.

[0103] Indeed, the brightness value may have a different influence depending on the different types of display devices, and the characteristics of the images to be displayed. The use for learning of a characteristic relating to the brightness of the screen, in combination with characteristics relating to the image and / or the digital display device therefore allows the supervised automatic learning engine to learn the existing correlations between these different characteristics for the prediction of the induced electrical consumption.

[0104] The characteristics may also include at least one characteristic of the environment of the digital display device.

[0105] The at least one characteristic of the environment of the device may for example comprise a position and / or an orientation of the digital display device. For example, the consumption may vary depending on the orientation of the device (for example with higher brightness if the device is oriented towards the south) or the position of the device (for example, depending on the presence of other neighboring devices). This makes it possible for example to predict the consumption induced by the display of images depending on the arrangement of the device.

[0106] The at least one characteristic of the environment of the device may also include at least one meteorological data item such as for example a temperature, humidity or ambient brightness value when displaying the image.

[0107] Indeed, depending on the technologies implemented by the display devices, these ambient values ​​can impact the quantity of electrical energy induced by the display of an image. For the constitution of the training base, the meteorological data values ​​can for example be obtained by means of meteorological predictions, or directly by means of sensors of the display devices.

[0108] This therefore makes it possible to predict the consumption induced by the display of an image as a function of the weather conditions expected during the display. In the case of characteristics of the environment of the device, the training can be done on the basis of the characteristics of the environment of the device alone, or of the characteristics of the environment of the device in connection with the characteristics of the device itself and / or the characteristics of the image, in order to learn the correlations existing between the weather data, the characteristics of the image and / or the display device, and the electricity consumption induced by the display of the image.

[0109] [Fig.3] represents a Sys3 system for displaying at least one image by at least one digital display device according to a set of embodiments of the invention.

[0110] The Sys3 system includes: - at least one digital display device equipped with a screen. In the example of [Fig.3] devices Disp31, Disp32 and Disp33, equipped respectively with 3 screens Ecr31, Ecr32 and Ecr33 are represented; - at least one computing device. In the example of [Fig.3], a server Serv3 is shown. The at least one computing device comprises: • a Mem3 memory storing a previously trained machine learning engine, for example by a method such as represented in figure 2; • at least one Calc3 calculation unit; - a Conn3 connection between the at least one computing device and the at least one digital display device.

[0111] [Fig. 3] is provided as a non-limiting example only of a Sys3 system for displaying at least one image by at least one digital display device according to a set of embodiments of the invention, and other variations are possible. For example: - the display devices shown in [Fig.3] are fixed devices, but a Sys3 system may also include mobile display devices such as smartphones or tablets; - in [Fig.3], the at least one computing device is a server. However, the invention is not limited to this example, and the Sys3 system may comprise a single device or a plurality of computing devices. The computing devices may be servers, or other computing devices such as personal computers; - if [Fig.3] shows a single computing unit Calc3 and a single memory Mem3, several computing units and memories can be distributed over one or more computing devices.

[0112] The at least one Calc3 calculation unit can be configured to predict the power consumption induced by the display of an image or content on the devices Disp31, Disp32 or Disp33, for example by executing a method such as represented in [Fig.4].

[0113] The at least one Calc3 calculation unit may also be configured to trigger the display of an image or content by a digital display device, for example by sending the image (or the content comprising the image) and a display instruction to the digital display device.

[0114] As will be described in more detail below, the at least one computing unit can make it possible to reduce the electrical consumption induced by the display of the images, for example in one of the following ways: - by modifying images / content to reduce the consumption induced by the display; - by selecting the display device from a set of candidate devices most suitable for displaying a given image / content, or conversely by selecting the content most suitable for a display device from a set of given contents; - by defining the placement parameters of a display device allowing an overall reduction in the electrical consumption induced by the display of images on the device.

[0115] Reference is now made to [Fig.4].

[0116] [Fig.4] shows a method P4 for predicting the electrical consumption induced by the display of at least one image by a digital display device according to one embodiment. The method P4 can for example be implemented by the at least one Calc3 calculation unit.

[0117] The method P4 comprises a first step of obtaining at least one characteristic belonging to a group of characteristics comprising: - characteristics of the digital display device; - characteristics of the environment of the digital display device; - characteristics of the at least one image to be displayed;

[0118] In essence, this step involves obtaining features corresponding to the expected features as input to the supervised machine learning engine and corresponding to the intended display. For example, obtaining the features may include: - obtaining the image to display; - the extraction of one or more characteristics representative of the image to be displayed; - obtaining one or more characteristics of the display device (e.g. model, manufacturer, screen type, etc.); - obtaining at least one characteristic of the device's environment: position, orientation, meteorological data measured by the display device or predicted, etc.

[0119] The method P4 then comprises a second step S42 of executing a supervised machine learning engine taking as input said at least one characteristic, the supervised machine learning engine having been previously trained, for example by a method such as represented in [Fig.2].

[0120] The method P4 thus makes it possible to predict the electrical consumption induced by the display of the image by the digital display device.

[0121] If an image is part of a larger content, for example a video, the overall power consumption of the content can also be predicted, for example by summing the predictions of the induced power consumption of each image of the content. Thus, the method P4 makes it possible to predict, for a given content, an overall value of power consumption induced by the display of the content.

[0122] The induced electricity consumption can also be converted into other induced values ​​depending on the electricity consumption. For example, the induced electricity consumption can be converted into induced greenhouse gas emissions, induced costs, etc.

[0123] Thus, the P4 method makes it possible to associate with a given content an induced electrical consumption, but also an associated carbon footprint or operating cost.

[0124] Predicting the electricity consumption induced by the display of an image makes it possible to reduce this electricity consumption in several ways. For example, content can be modified so that these images induce lower electricity consumption, content can be broadcast on a digital display device inducing lower electricity consumption for the images of this content, a digital display device can be implemented so that the electricity consumption generally induced by the broadcast images is lower, etc.

[0125] The P4 process therefore ultimately makes it possible to reduce the electricity consumption induced by the display of content, but also the carbon footprint associated with the content or even the operating cost linked to the display of the content.

[0126] Reference is now made to Figures 5a, 5b, 6a and 6b.

[0127] Figures 5a and 5b show respectively: - a GraphMes5 graph representing the electrical consumption induced by the display of a first content on a first digital display device, as measured during the display of the first content on the first digital display device; - a Pred5 graph representing the electrical consumption induced by the display of the first content on the first digital display device, as predicted by a method in an embodiment of the invention, for example the method P4.

[0128] Figures 6a and 6b show respectively: - a GraphMesô graph representing the electrical consumption induced by the display of a second content on a second digital display device, as measured during the display of the second content on the second digital display device; - a Pred6 graph representing the electrical consumption induced by the display of the second content on the second digital display device, as predicted by a method in an embodiment of the invention, for example the method P4.

[0129] Each of the graphs GrphMes5, GrphPrd5, GraphMesô and GrphPrdô shows: - on the x-axis, the frame number of each image of the first or second content. For example, we note here that the first content includes 253 images and the second content includes 365; - on the y-axis, the electrical consumption induced (predicted or measured) by the display.

[0130] The curves Mes5, Pred5, Mes6 and Pred6 therefore represent the measurements or predictions of the electrical consumption induced by the display of successive images of each content.

[0131] We observe on these curves that the electrical consumption induced by the display of the images is very different depending on the images. We also note that the prediction turns out to be quite close to the measurements.

[0132] These examples demonstrate the ability of the invention to reliably predict the electricity consumption induced by the display of images on digital display devices. These examples also show that the electricity consumption induced by the display of images can be very different depending on the images. The ability to predict this induced electricity consumption therefore makes it possible to reduce it significantly, for example according to one of the embodiments described in FIGS. 7 to 9.

[0133] Reference is now made to [Fig.7].

[0134] [Fig.7] shows an example of a method for predicting the electrical consumption induced by the display of images on a digital display device of a device as a function of at least one characteristic of the device or its environment according to one embodiment.

[0135] Method P7 comprises all the steps of method P4.

[0136] In the example of [Fig.7], the at least one characteristic comprises at least one characteristic of the digital display device or the environment of the device.

[0137] For example, the at least one feature may include at least one of the following features: - Device characteristics: • Screen type; • Screen size; • Manufacturer; • Etc. - characteristic of the device environment: • position ; • orientation ; • average ambient brightness; • average temperature; • average humidity level; • etc.

[0138] The method P7 also comprises a step S74 of modifying said at least one characteristic of the digital display device to reduce the prediction of the electrical consumption induced by the display of the at least one image.

[0139] For example, the modification of the characteristics can be done iteratively: a criterion for the end of the modifications can be tested at a step S73, then: - if the criterion is not verified, the modification step S74 is executed, followed by a new prediction step S42; - if the criterion is verified, the modification is completed, and subsequent steps, such as for example a step S75 of device deployment, can be executed.

[0140] The criterion for ending the modifications in step S73 may be of different types. For example, it may be a criterion relating to a number of limit iterations, obtaining a given level of reduction in the induced consumption, a convergence criterion verified when the consumption no longer decreases after several iterations, etc.

[0141] In other words, the method P7 may correspond to an iterative modification of at least one parameter of a digital display device or its environment to reduce the electrical consumption induced by the display of images on the device.

[0142] For example, the method P7 can optimize the position, orientation or even the parameters of the device (for example the type of screen or fan) in order to minimize the electrical consumption generally induced by the display of images on the device.

[0143] The P7 method therefore makes it possible to obtain more economical digital display devices.

[0144] The optimization can be done on the basis of a single image, or on the basis of a set of images. For example, the prediction in step S42 can be carried out on a set of images of a test base, and the average induced consumption can be retained.

[0145] The modifications in step S74 can also be made in different ways. For example, a set of possible modifications (e.g. a set of possible positions and orientations around a reference position and orientation, a set of possible characteristics of the display device, etc.) can be exhaustively tested to retain the best one. The choice of modifications to be tested can also be made by means of optimization algorithms, such as for example genetic or gradient descent algorithms.

[0146] As indicated above, once the modifications to the characteristics are completed, one or more subsequent steps may be performed.

[0147] For example, the method P7 may conclude with a step S75 of deploying the device according to at least one modified characteristic. For example, this step may consist of physically installing the digital display device according to the position, orientation or characteristics of the device as modified in the previous steps.

[0148] Reference is now made to [Fig.8].

[0149] [Fig.8] shows an example of a method for modifying at least one characteristic of at least one image to reduce the electrical consumption induced by the display of the at least one image, according to a set of embodiments of the invention.

[0150] Method P8 comprises all the steps of method P4.

[0151] In the context of the method P8, the at least one characteristic comprises at least one characteristic of the at least one image to be displayed. For example, the at least one characteristic may comprise the image itself, or characteristics representative of the image such as the average brightness, the contrast, the luminance values ​​for each pixel, etc.

[0152] The method P8 also comprises a step S84 of modifying said at least one characteristic of the digital image to reduce the prediction of the electrical consumption induced by the display of the at least one image.

[0153] For example, the modification of the characteristics can be done iteratively: a criterion for the end of the modifications can be tested at a step S83, then: - if the criterion is not verified, the modification step S84 is executed, followed by a new prediction step S42; - if the criterion is verified, the modification is completed, and subsequent steps, such as for example a step S85 of displaying the at least one modified image.

[0154] The criterion for ending the modifications in step S83 may be of different types. For example, it may be a criterion relating to a number of limit iterations, obtaining a given n level of reduction in the induced consumption, a convergence criterion verified when the consumption no longer decreases after several iterations, etc.

[0155] In other words, the method P8 can correspond to an iterative modification of at least one parameter of an image (for example, of the image itself or of a parameter representative of the image such as the average brightness, the contrast of the image, etc.).

[0156] For example, the method P8 can optimize the image, either by means of direct modifications of the light intensity values, or by means of modifications of global parameters of the image (brightness, contrast, etc.) in order to minimize the electrical consumption generally induced by the display of the at least one image.

[0157] The P8 method therefore makes it possible to obtain images whose display induces lower electrical consumption.

[0158] The optimization can be done on the basis of a single image, or on the basis of a set of images. For example, the prediction in step S42 and the modification in step S74 can be carried out on all the images of a given content, for example a video.

[0159] The prediction in step S42 can also be made either for a given particular device (for example the device on which the at least one image must be displayed), or for a plurality of devices. For example, a prediction of the electrical consumption induced by the display of an image can be made for each device of a set of reference digital display devices, and the average of the consumptions induced for the different devices can be retained. Thus, this makes it possible to estimate the consumption induced by the display of the image on a set of possible target display devices.

[0160] Similarly, the consumption prediction can be done either for a single image or for a set of images, for example all the images of a content, for example a video. In this second case, the sum or the average of the consumptions can be taken into account, in order to reduce the consumption induced by the display of a content.

[0161] The modifications in step S84 can also be made in different ways. For example, the modification can be made manually by a user to whom the induced consumptions are presented at each iteration. Thus, the user can test different modifications in order to reduce the induced consumption. For example, a prediction curve of the consumptions induced by the images of a content such as the Pred5 or Pred6 curves can be presented to the user, who can then select one or more images to modify (for example the most consuming ones), then make the modifications and restart the prediction, and so on to reduce the consumption induced by the display of the content.

[0162] Modifications can also be made automatically. For example, a set of possible modifications (e.g., smoothing an image, reducing brightness or contrast, etc.) can be tested exhaustively to select the best one. The choice of modifications to be tested can can also be done through optimization algorithms, such as genetic algorithms or gradient descent.

[0163] As indicated above, once the modifications to the characteristics are completed, one or more subsequent steps may be performed.

[0164] For example, the method P8 may conclude with a step S85 of displaying the at least one modified image on a digital display device. For example, the modified content may be broadcast on one or more digital display devices. The electrical consumption induced by the broadcasting of the content will thus be reduced.

[0165] The P8 method therefore allows manual or automatic modification of images to display content while limiting the associated electricity consumption. In fact, the costs and greenhouse gas emissions induced by the distribution of content are also reduced.

[0166] Reference is now made to [Fig.9].

[0167] [Fig.9] shows an example of a method P9 for selecting content to be displayed according to one embodiment.

[0168] Method P9 comprises all the steps of method P4.

[0169] In the example of [Fig.9], the prediction in step S42 is carried out for a plurality of distinct contents. For example, the method P9 may comprise a step S93 of verifying that another content is to be tested, and: - if another content is to be tested, a return to step S42 of prediction of the induced electrical consumption; - otherwise, a content selection step S94 is activated.

[0170] The contents may comprise either a single image or a plurality of images. In this second case, the prediction in step S42 may be performed for each image, and an overall induced electrical consumption may be determined for the content, for example as the sum or the average of the consumptions induced by the different images of the content.

[0171] Thus, at the end of all the iterations of steps S42 and S93, each content is associated with a prediction of an electrical consumption induced by the display of the content.

[0172] The method P9 then comprises a step S94 of selecting at least one content to be displayed from among said plurality of contents, in order to minimize the prediction of the electrical consumption induced by the display of the at least one content.

[0173] In other words, step S94 consists of selecting a content from the set of tested contents, which makes it possible to minimize the electrical consumption induced by the display. For example, the content associated with the lowest induced electrical consumption prediction can be selected.

[0174] The method P9 may further also comprise a subsequent step of displaying the at least one selected content on the at least one digital display device.

[0175] Thus, the method P9 makes it possible to select, and where appropriate to display, content from a set of given contents which minimizes the electrical consumption induced by the display.

[0176] The P9 method therefore makes it possible to reduce the electricity consumption induced by the display of content on digital screens. It therefore also makes it possible to reduce the consumption of greenhouse gases induced by the display, and the cost of the display.

[0177] Reference is now made to [Fig. 10].

[0178] [Fig. 10] shows an example of a method P10 for selecting a device on which to display content according to one embodiment.

[0179] Method P10 comprises all the steps of method P4.

[0180] In the example of [Fig. 10], the prediction in step S42 is performed for a plurality of distinct display devices. For example, the method P10 may comprise a step S103 of verifying that another display device is to be tested, and: - if another display device is to be tested, a return to step S42 of prediction of the induced electrical consumption; - otherwise, a step S104 for selecting the display device is activated.

[0181] In other words, the power consumption induced by the display of content is predicted for the display of the content on a plurality of distinct display devices.

[0182] The contents may comprise either a single image or a plurality of images. In this second case, the prediction in step S42 may be performed for each image, and an overall induced electrical consumption may be determined for the content, for example as the sum or the average of the consumptions induced by the different images of the content.

[0183] Thus, at the end of all the iterations of steps S42 and S103, each display device is associated with a prediction of an electrical consumption induced by the display of the content.

[0184] The method P10 then comprises a step S104 of selecting at least one digital display device on which to display the image from among said plurality of digital display devices, in order to minimize the prediction of the electrical consumption induced by the display of the at least one image.

[0185] In other words, step S104 consists of selecting from the set of digital display devices tested, the one which makes it possible to minimize the consumption induced electricity consumption by displaying the content. For example, the digital display device associated with the lowest induced electricity consumption prediction can be selected.

[0186] The method P10 may further also comprise a subsequent step of displaying the at least one image on said selected digital display device.

[0187] Thus, the method P10 makes it possible to select a display device from a given set of display devices which minimizes the electrical consumption induced by the display of the content, and where appropriate to display the content on the selected device.

[0188] The P10 method therefore makes it possible to reduce the electricity consumption induced by the display of content on digital screens. It therefore also makes it possible to reduce the consumption of greenhouse gases induced by the display, and the cost of the display.

[0189] The present disclosure is not limited to the examples of method, system, and supervised machine learning engine described above, only by way of example, but it encompasses all the variants that may be envisaged by those skilled in the art within the framework of the protection sought.

Claims

Claims

1. Method (P2) for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device, said method comprising: - obtaining (S21) a training base from a plurality of training images, each sample of said training base comprising: • at least one characteristic belonging to a group of characteristics comprising: • characteristics of a digital display device; • characteristics of the environment of the digital display device; • characteristics of a training image; • a single consumption value induced by the display of said training image, said single induced power consumption value being obtained by means of a measurement carried out during the display of the training image;- training (S22) said supervised automatic learning engine on said training base to minimize a loss function between the induced electrical consumption predictions made by the supervised automatic learning engine and the induced consumption values ​​obtained using the measurements made when displaying the images.;

2. The method of claim 1 wherein the at least one feature comprises the training image.

3. The method of claim 2, wherein the at least one characteristic further comprises at least one characteristic of the digital display device on which the image is displayed.

4. A method according to any preceding claim, wherein the at least one characteristic comprises a value of brightness of a digital display screen on which the image is displayed at the time the image is displayed.

5. A method according to any preceding claim, wherein the at least one characteristic comprises at least one characteristic of the environment of the digital display device.

6. Method according to the preceding claim, in which the at least one characteristic of the environment of the digital display device comprises at least one meteorological data.

7. Method (P4; P7; P8; P9) for predicting the electrical consumption induced by the display of at least one image by a digital display device, said method comprising: - obtaining (S41) at least one characteristic belonging to a group of characteristics comprising: • characteristics of the digital display device; • characteristics of the environment of the digital display device; • characteristics of the at least one image to be displayed; - executing (S42) a supervised machine learning engine taking as input said at least one characteristic, said supervised machine learning engine having been previously trained by a method according to one of claims 1 to 6.

8. Method (P7) according to claim 7, wherein: - the at least one characteristic comprises at least one characteristic of the digital display device or of the environment of the device, - the method comprises a modification (S74) of said at least one characteristic of the digital display device or of the environment of the device to reduce the prediction of the electrical consumption induced by the display of the at least one image.

9. A method according to the preceding claim, further comprising deploying (S75) the device according to said at least one modified characteristic.

10. Method (P8) according to claim 7, wherein: - At least one characteristic comprises at least one characteristic of the at least one image to be displayed; - method further comprises modifying (S84) said at least one characteristic of the at least one image to be displayed to reduce the prediction of the power consumption induced by the display of the at least one image.

11. The method of the preceding claim, further comprising displaying (S85) the at least one modified image on a digital display device.

12. Method (P9) according to claim 7, wherein the method comprises: - predicting the electrical consumption induced by the display of a plurality of contents comprising at least one image by at least one digital display device; - selecting (S94) at least one content to be displayed from among said plurality of contents, in order to minimize the prediction of the electrical consumption induced by the display of the at least one content.

13. The method of claim 7, wherein the method comprises: - predicting the power consumption induced by the display of at least one image by a plurality of digital display devices; - selecting (S104) at least one digital display device on which to display the image from among said plurality of digital display devices, in order to minimize the prediction of the power consumption induced by the display of the at least one image.

14. System (Sysl) for training a supervised machine learning engine for predicting the power consumption induced by the display of at least one image by a digital display device, said system comprising:

15.

16.

17.

18. - at least one digital display device (Displ 1, Displ 2, Displ 3) capable of measuring electrical consumption induced by the display of at least one image; - at least one calculation unit (Calcl) configured to execute a method according to one of claims 1 to 6. System for displaying at least one image, said system comprising: - at least one digital display device; - at least one computing unit configured to execute a method according to one of claims 7 to 13. Deployed digital display device according to claim 9. Computer program comprising instructions for implementing the method according to one of claims 1 to 13 when this program is executed by a processor. Non-transitory recording medium readable by a computer on which is recorded a program for implementing the method according to one of claims 1 to 13 when this program is executed by a processor.

Images