Method and system of selective intelligent masking of images used as a proof of a traffic offence

Abstract

The invention relates to a selective masking and processing module (12) designed to cooperate with a system for detecting and proving a first type of traffic offences (11) to detect and prove one or more additional types of traffic offences. The system for detecting and proving a first type of traffic offences (11) includes : one or more image and / or video capture system(s) (111), optionally one or more violation detection sensor(s) (112) and a processing module (113) programmed to process the image (s) and / or video (s) provided by the image and / or video capture system(s) (111) and, if present, data and / or signals received from the violation detection sensor (s) (112), in order to detect any offences belonging to the first type of traffic offences, wherein the detected traffic offences are shown / associated in / with specific images selected from those provided by the image and / or video capture system (s) (111). The processing and selective masking module (12) is programmed to: implement a first module (41) that is configured to locate a front area of a vehicle present in a specific image, wherein said front area corresponds to a windscreen of the vehicle; apply the selective masking to a portion of the specific image that corresponds to a portion of the windscreen and on the basis of offences belonging to the first type of traffic offences detected by the processing module (113); and provide as output the specific image to which selective masking has been applied.

Description

[0001] "METHOD AND SYSTEM OF SELECTIVE INTELLIGENT MASKING OF IMAGES

[0002] USED AS A PROOF OF A TRAFFIC OFFENCE"

[0003] Cross-Reference to Related Applications

[0004] This Patent Appl ication claims priority from European Patent Application No . 23425033 . 0 filed on June 16 , 2023 and Italian Patent Application No . 102023000019356 filed on September 20 , 2023 , the entire disclosure of which is incorporated herein by reference .

[0005] Technical Field of the Invention

[0006] The present invention relates to a method and a related system for the selective intelligent masking of images used as proof of a traf fic of fence .

[0007] State of the Art

[0008] Electronic systems for the automated detection of traf fic of fences , similarly to video surveillance systems , involve the processing of personal data . Thus , the use of such systems is only allowed for the collection of data strictly relevant to the detected of fence , whereby the location and angle of view o f the shooting are limited so as not to collect irrelevant and / or unnecessarily detailed images . In compliance with the practice and the regulatory framework of the sector regarding certain traf fic violations , the data protection authority, Garante per la Protezione dei Dati Personal! ( GPDP ) , establishes the following :

[0009] • electronic detection systems must restrict the storage of alphanumeric data contained in vehicle number plates only to cases wherein road traf fic provisions are not ful filled; • the photographic findings and / or video shooting can only identi fy the elements provided for by the sector regulations for preparing the report ascertaining the violations ( e . g . , pursuant to Art . 383 of Presidential Decree no . 495 / 1992 , the type of vehicle , the day, the time and the place where the violation took place ) , wherein a shooting of the vehicle must be made that does not include or, alternatively, masks , as far as possible , the portion of the video and / or photographic findings concerning persons not involved in the administrative check, for example pedestrians and / or other road users ;

[0010] • the photographic findings and / or video shooting obtained are to be used only to ascertain violations of road traf fic provisions also during a dispute , without prej udice to their accessibility by those entitled to them;

[0011] • the images must be stored for the period of time strictly necessary in relation to the dispute , to the possible application of a sanction and to the settlement of the possible litigation in accordance with the relevant legislation, without prej udice to any further storage requirements arising from a speci fic investigative request by the j udicial authorities and / or the police ;

[0012] • photographs and / or images constituting a source of evidence for disputed violations are not to be sent to the vehicle holder ' s home address together with the report , without prej udice to their accessibility to those entitled to them; and

[0013] • in view of the legitimate interest of the holder of the vehicle to veri fy the author of the violation and, therefore, to obtain from the competent authority any useful element for this purpose, the viewing of the documentation (i.e. the photographic findings and / or video shooting) must be made available upon request of the addressee of the report; moreover, at the time of access, the passengers on board the vehicle must be suitably obscured or in any case made unrecognisable.

[0014] In this context, the rule thus consists in the following two principles:

[0015] • the privacy of the road user must be protected; and

[0016] • protection is not due for the person who is responsible for a violation.

[0017] In order to comply with these provisions, in Italy, for example, the use of electronic systems that shoot vehicles committing violation from the rear is preferred, so that one does not have to worry about the privacy of those on board as these persons are not shot.

[0018] The most advanced electronic systems (e.g. speed detection systems) , also as a result of appropriate circulars, e.g. by the Ministry of the Interior, have developed solutions that have been approved for the automatic detection of violations (e.g. by the Ministry of Infrastructure and Transport) and that allow, in case the vehicle is shot from the front, to automatically obscure the windscreen of the vehicle committing violation, irreversibly masking the image of the vehicle committing violation captured by said electronic systems and making the persons inside the vehicle unrecognisable.

[0019] Figure 1 shows an example of an image captured by a speed detection system with a vehicle shot from the front and the front part of said vehicles masked; this mode of operation, however, implies a number of negative consequences , first and foremost the impossibility of recognising the driver of the vehicle ; in fact , the driver of the vehicle violating the traf fic code should be identi fiable with no need to protect his or her privacy . On the other hand, in many cases , people who committed violations declare that they do not remember who was driving and manage , with a simple additional fine , to avoid the possible accessory sanctions provided for ( e . g . reduction of points from the driving licence ) , as well as possible of fences in the event of recidivism.

[0020] In this context , DE 102016221521A1 relates to a method for masking an image of a vehicle taken by a traf fic monitoring device . In particular, the information representing the image is analysed in order to mark at least one image section and mask the marked image section .

[0021] Object and Summary of the Invention

[0022] The aim of the present invention is to provide a technical solution that is able to overcome , or at least alleviate , the technical disadvantages of currently known solutions for automated traf fic of fence detection .

[0023] This and other aims are achieved by the present invention as it relates to a selective masking and processing method, as defined in the attached claims .

[0024] In particular, the selective masking and processing module is designed to cooperate with a system for detecting and proving a first type of traf fic of fences to detect and prove one or more additional types of traffic of fences . The system for detecting and proving a first type of traffic offence includes one or more image and / or video capture systems, optionally one or more violation detection sensors and a processing module programmed to process the image (s) and / or video (s) provided by the image and / or video capture system(s) and, if present, data and / or signals received from the violation detection sensor (s) , in order to detect any offences belonging to the first type of traffic offences, wherein the detected traffic offences are shown / associated in / to specific images selected from those provided by the image and / or video capture system(s) . The selective masking and processing module is programmed to implement a first module that is configured to locate a front area of a vehicle in a specific image, wherein said front area corresponds to a windscreen of the vehicle; to apply the selective masking to a portion of the specific image that corresponds to a portion of the windscreen and based on the offences belonging to the first type of traffic offences detected by the processing module; and to output the specific image to which selective masking has been applied.

[0025] Brief Description of the Drawings

[0026] For a better understanding of the present invention, some preferred embodiments (provided purely by way of explanatory, but absolutely not limiting, let alone binding example) will now be shown with reference to the accompanying drawings (not to scale) , wherein:

[0027] • Figure 1 shows an example of an image captured by a speed detection system with a vehicle shot from the front according to the prior art; • Figures 2-7 show examples of images captured by implementing a method for detecting and proving traf fic of fences according to the present invention;

[0028] • Figures 8- 10 schematically shows examples of systems architecture according to preferred embodiments of the invention; and

[0029] • Figure 11 schematically shows an example of the operating logic of a selective masking and processing module of the systems shown in Figures 8- 10 .

[0030] Description of the Invention

[0031] The following description is provided to enable a person skilled in the art to make and use the invention . Various modi fications to the embodiments set forth will be immediately clear to the persons skilled in the art and the general principles herein disclosed may be applied to other embodiments and applications without , however, departing from the protection scope of the present invention as defined in the appended claims .

[0032] Therefore , the present invention should not be understood as limited to the sole embodiments described and shown, but it must be given the widest scope of protection in accordance with the characteristics defined in the appended claims .

[0033] The present invention relates to a method and system for detecting and proving traf fic of fences , wherein said method and system involve processing images captured by a system for detecting and proving a first type of traf fic of fences by selectively masking the front part of the vehicles shown in the images so as to enable recognition / identi f ication of the person driving and to prove any further violations committed by him / her, and possibly the recognition / identif ication of other persons in the vehicles committing further types of traffic offences for which, therefore, no privacy is to be protected.

[0034] In particular, the present invention involves the use of a selective masking and processing module, preferably based on artificial intelligence algorithms, of images captured by a system for detecting and proving a first type of traffic offences (e.g. exceeding the speed limit, detection of wrongway vehicles, prohibited overtaking, etc.) , wherein said selective masking and processing module is able to detect any further types of offences committed by persons on board a vehicle shot in an image (e.g., seat belt not fastened, use of the telephone while driving, etc.) by implementing a corresponding selective masking of the image.

[0035] Figures 2-7 show examples of traffic offences that can be detected by using the system, i.e. by implementing the method, according to the present invention, wherein:

[0036] • Figure 2 shows a first image of the front of a first vehicle, said first image being only related to a speed limit violation, so only the driver of the first vehicle is visible, while a possible passenger is not visible through selective masking;

[0037] • Figure 3 shows a second image of the front of a second vehicle, said second image being related to both a speed limit violation and the failure of the driver of the second vehicle to fasten the seat belt, so that only the driver is visible, while a possible passenger is not visible through selective masking;

[0038] • Figure 4 shows a third image of the front of a third vehicle , said third image being related both to a speed limit violation and to the use of the telephone while driving by the driver of the third vehicle , so that only the driver is visible , while a possible passenger is not visible through selective masking;

[0039] • Figure 5 shows a fourth image of the front of a fourth vehicle , said fourth image being related both to a speed limit violation, to the use of the telephone while driving by the driver of the fourth vehicle , and to the failure of the driver to fasten the seat belt , so that only the driver is visible , while a possible passenger is not visible through selective masking;

[0040] • Figure 6 shows a fi fth image of the front of a f i fth vehicle , said fi fth image being related to both a speed limit violation and to the failure of the driver and a passenger of the fi fth vehicle to fasten the seat belt , so that both driver and passenger are visible ;

[0041] • Figure 7 shows a sixth image of the front of a s ixth vehicle , said sixth image being related both to a speed limit violation, to the presence of two passengers , instead of j ust one as required by the law for that type of vehicle , next to the driver of said sixth vehicle , and to the failure of the driver and the two passengers to fasten their seat belts , so that both the driver and the two passengers are visible .

[0042] The present invention is intended to prove possible traf fic violations by implementing selective masking adapted to show only those persons who commit offences and to make those who do not commit violations non-visible. For example, if a person has exceeded the speed limit and further violations can be detected from the images proving his or her violation (e.g. not having the seat belt fastened and / or using a mobile phone) , the law allows for the application of the foreseen penalties.

[0043] As shown in Figures 6 and 7, a selective masking algorithm is used on the image captured to prove a violation belonging to those that the law allows to be automatically detected, making it possible not to mask the portion of the image showing the passenger when the latter is not complying with the traffic code provisions, for example when he / she has not the seat belt fastened.

[0044] Figures 8-10 schematically show examples of systems architecture according to preferred embodiments of the invention .

[0045] In particular, Figure 8 shows a system for detecting and proving traffic offences (overall denoted by 1) that includes :

[0046] • a system for detecting and proving a first type of traffic offences 11 (e.g. exceeding speed limit, detection of wrong-way vehicles, prohibited overtaking, etc.) that includes

[0047] - one or more image and / or video capture systems (111) ,

[0048] - optionally one or more sensors 112,

[0049] - a processing module 113 programmed to process the image (s) and / or video (s) provided by the image and / or video capture system(s) 111 and, if applicable, the data provided by the sensor (s) 112, in order to detect any offences belonging to the first type of traffic offences, wherein the detected offences are shown / associated in / to specific images selected from those provided by the image and / or video capture system (s) 111; and

[0050] • a selective masking and processing module 12 programmed to

[0051] - process the specific images provided by the processing module (113) in order to detect any additional types of traffic offences shown in said specific images,

[0052] - apply selective masking to specific images on the basis of

[0053] - offences belonging to the first type of traffic offences detected by the processing module 113, and

[0054] - any additional types of traffic offences detected by said selective masking and processing module 12; and

[0055] - provide as output the specific images to which selective masking has been applied.

[0056] Figure 9 shows a system for detecting and proving traffic offences comprising the system for detecting and proving a first type of traffic offences 11 and the selective masking and processing module 12, wherein said selective masking and processing module 12 processes the proof of violations as issued by the system 11 (i.e., images and / or videos) and transmits the specific images and / or videos to which selective masking has been applied to a remote system 2 via wired or wireless transmission means 3 .

[0057] Figure 10 shows the detection system for a first type of traf fic of fences 11 that transmits , via wired or wireless transmission means 4 , the speci fic images to the selective masking and processing module 12 that is integrated into a remote system 5 .

[0058] From an architectural perspective , as shown in Figures 8- 10 , the selective masking and processing module 12 can reside within a generic system for detecting and proving traf fic of fences 1 ( Fig . 8 ) . It is also possible that the selective masking and processing module 12 can be implemented outside the system for detecting and proving a first type of traf fic of fences 11 either through an appropriate local processing system ( Fig . 9 ) , or within a remote processing system 5 dedicated to other activities ( Fig . 10 ) .

[0059] Figure 11 schematically shows an example of the operation logic of the selective masking and processing module 12 by implementing a plurality of software modules .

[0060] The selective masking and processing module 12 receives as input an image and processes it , preferably by means of arti ficial intelligence algorithms based on obj ect detection, evaluating whether said input image is to be selectively masked or not . In particular, the front area of the vehicle in the image is located on the input image , wherein this front area of the vehicle generally coincides with the windscreen of the vehicle (block 41 in Figure 11 ) .

[0061] The definition of the area to be masked or not of the vehicle in the input image depends on the presence of a module for detecting and locating the presence of one or more passenger (s) being carried (block 42 in Figure 11) which, referring to the vehicle driving position (e.g. whether driving is on the left or on the right) , detects the presence of the passenger (s) being carried and, in this case, locates said carried passenger ( s ) ; in this regard, it is worth noting that the front area of the vehicle corresponding to the driver's position is never masked as the need to identify the person committing violation (e.g. for exceeding the speed limit) obviates the need to protect his or her privacy .

[0062] On the other hand, in the event that the presence detection and location module of one or more carried passengers (block 42 in Figure 11) is not present, given that the driving position is known (i.e., left or right) , the position of the carried passenger (s) can be detected on the input image (which will be left or right respectively) , whereby the corresponding portion of the input image is masked (block 44 in Figure 11) .

[0063] If the presence detection and location module of one or more carried passenger (s) is present (block 42) , if the carried passenger (s) is / are not present inside the vehicle in the input image, then there is no passenger being carried whose privacy needs to be protected and, therefore, there is no need for masking; in this case, the output image corresponds to the input image, i.e. to the same image captured by the system for detecting and proving a first type of traffic offences 11.

[0064] Optionally, a violation detection module of the passenger (s) being carried can also be implemented (block 43 in Figure 11) . In this case, if the passenger (s) being carried is present, it is possible to detect any violations committed by said passenger (s) being carried (e.g. failure to use the seat belt) . As shown in Figure 11, if the passenger (s) being carried is / are in violation, there is no need for masking, and as there is no person whose privacy needs to be protected, there is no need for masking, issuing the output image, which will correspond to the input image.

[0065] By contrast, in the event that the violation detection module of the carried passenger (s) (block 43) and / or the presence and location detection module of one or more carried passenger (s) (block 42) is / are not present and / or the carried passenger (s) is / are not in violation, the input image portion corresponding to the carried passenger (s) (block 44) is masked in order to protect his or her privacy by issuing the output image that has the portion of the windscreen corresponding to the passenger side masked.

[0066] From the foregoing disclosure, the many innovative characteristics and the innumerable technical advantages of the present invention are immediately evident to a person skilled in the art.

[0067] In conclusion, it is important to note that, while the above-described invention refers in particular to very specific embodiments, it must not be intended as limited to such embodiments, including within its scope all the variants, modifications or simplifications covered by the appended claims.

[0068] In this regard, it is important to draw the attention to the fact that the selective masking and processing module 12 described in detail above can be conveniently implemented by an appropriate computer programme (i.e., software and / or firmware) , or by means of a suitably programmed device / system.

Claims

CLAIMS1. Selective masking and processing module (12) designed to cooperate with a system for detecting and proving a first type of traffic offences (11) to detect and prove one or more additional types of traffic offences; wherein the system for detecting and proving a first type of traffic offences (11) includes:• one or more image and / or video capture systems (111) ;• optionally one or more violation detection sensors (112) ; and• a processing module (113) programmed to process the image (s) and / or video (s) provided by the image and / or video capture system(s) (111) and, if present, data and / or signal (s) received from the violation detection sensor (s) (112) , in order to detect any offences belonging to the first type of traffic offences, wherein the detected traffic offences are shown / associated in / with specific images selected from those provided by the image and / or video capture system(s) (111) ; wherein the selective masking and processing module (12) is programmed to:• implement a first module (41) that is configured to locate a front area of a vehicle present in a specific image, wherein said front area corresponds to a windscreen of the vehicle ;• apply a selective masking to a portion of the specific image that corresponds to a portion of the windscreen and based on the offences belonging to first type of traffic offences detected by the processing module (113) ; and• provide as output the specific image to which the selective masking has been applied.

2. The module of claim 1, wherein the selective masking and processing module (12) is programmed to:• process the specific images provided by the processing module (113) in order to detect any additional types of traffic offences shown in said specific images; and• apply a selective masking to the specific images on the basis of both the offences belonging to the first type of traffic offences detected by the processing module (113) and any further types of traffic offences detected by said selective masking and processing module (12) .

3. The module of claim 1 or 2, configured to process the specific images and apply selective masking to said specific images by implementing one or more artificial intelligencebased algorithms, preferably one or more artificial intelligence techniques for object detection.

4. The module according to any one of claims 1-3, programmed to apply the selective masking to one or more of the specific images by applying a selective masking to a portion of the specific image that corresponds to a portion of the windscreen on the passenger side.

5. The module according to any one of claims 1-3, programmed to apply selective masking to one or more of the specific images by implementing a second module (42) that is configured to detect the presence of, and locate, one or more passengers in the front area of the vehicle; wherein, if the second module (42) does not detect the presence of passengers, the selective masking and processingand module (12) is programmed to not apply any selective masking to the specific image; and wherein, if the second module (42) detects the presence of the one or more passengers, the selective masking and processing module (12) is programmed to apply the selective masking to a portion of the specific image that corresponds to a portion of the windscreen on the passenger side.

6. The module according to any one of claims 1-3, programmed to apply the selective masking to one of the specific images by implementing:• a second module (42) that is configured to detect the presence of, and locate, one or more passengers in the front area of the vehicle; and• a third module (43) which is configured to detect any offences by one or more passengers; wherein, if the second module (42) does not detect the presence of passengers, the selective masking and processing module (12) is programmed to not apply any selective masking to the specific image; wherein, if the second module (42) detects the presence of the one or more passengers and the third module (43) detects one or more offences, the selective masking and processing module (12) is programmed to not apply any selective masking to the specific image; wherein, if the second module (42) detects the presence of the one or more passengers and the third module (43) does not detect any offences, the selective masking and processing module (12) is programmed to apply the selective masking to a portion of the specific image that corresponds to a portionof the windscreen on the passenger side.

7. The module according to any preceding claim, integrated into a system (1) for detecting and proving one or more types of traffic offences together with the system for detecting and proving a first type of traffic offences (11) , or implemented by a processing device / system coupled / connected to said system for detecting and proving a first type of traffic offences (11) .

8. The module according to any claim 1-6, integrated into a remote system (5) .

9. The module of claim 8, wherein the remote system (5) is remotely connected to the system for detecting and proving a first type of traffic offences (11) .

10. Electronic processing apparatus programmed as the selective masking and processing module (12) as claimed in any preceding claim.

11. Computer program product comprising firmware and / or software code portions that are:• loadable / storable on an electronic processing device / system; and• such that to cause, when loaded / stored, said electronic processing device / system to become programmed as the selective masking and processing module (12) as claimed in any one of claims 1-9.

Images