DE212018000230U1 - Pedestrian dynamics display system - Google Patents

Abstract

A system for displaying pedestrian dynamics, comprising: omnidirectional cameras configured to acquire N dynamic pedestrian video signals and to transmit the respective multiple pedestrian dynamic video signals to a central processor, where N is a positive integer; a central processor configured to splicing and compressing the N dynamic pedestrian video signals and sending the video data obtained after splicing and compression to a display screen; a display screen configured to decode the spliced and compressed video data to obtain decoded video data and on which the decoded video data is displayed to display pedestrian dynamics on the display screen.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

The present application claims priority from Chinese Patent Application No. CN201710376818.7 entitled "Method and System for Displaying Pedestrian Dynamics", filed May 25, 2017, the entire contents of which are incorporated herein by reference.

TECHNICAL AREA

The present invention relates to the field of traffic monitoring and multimedia technology, and more particularly to a system for displaying pedestrian dynamics.

STATE OF THE ART

In the prior art, traffic lights are set up in places where there are many pedestrians in order to regulate the behavior of pedestrians and cyclists and to avoid traffic accidents. As a rule, however, no traffic lights are placed under a ramp, since the danger increases when there is a traffic light under a ramp and a vehicle is parked on the ramp. On the other hand, without the traffic lights, the cyclist cannot know the pedestrian dynamics under the ramp in real time, which is why traffic safety cannot be guaranteed.

Therefore, there is a problem in the prior art that the cyclist does not know the pedestrian dynamics under the ramp in real time and that traffic safety cannot be guaranteed.

CONTENT OF THE PRESENT INVENTION

The main object of the present invention is to propose a system for displaying the pedestrian dynamics in order to solve the problem that the cyclist does not know the pedestrian dynamics under the ramp in real time and the traffic safety cannot be guaranteed.

• Erfassen von N dynamischen Videosignalen der Fußgänger und Senden der jeweils erfassten mehreren dynamischen Videosignalen der Fußgänger an einen Zentralprozessor durch N omnidirektionale Kameras, wobei N eine positive ganze Zahl ist;
• Spleißen sowie Komprimieren der N dynamischen Videosignalen der Fußgänger und Senden der nach dem Spleißen sowie Komprimieren erhaltenen Videodaten an einen Anzeigebildschirm durch den Zentralprozessor;
• Decodieren der gespleißten sowie komprimierten Videodaten durch den Anzeigebildschirm, um decodierte Videodaten zu erhalten, und Anzeigen der decodierten Videodaten auf dem Anzeigebildschirm, um die Fußgängerdynamik auf dem Anzeigebildschirm anzuzeigen.

To achieve the above object, the present invention provides a system for displaying pedestrian dynamics, the system comprising the following functions:

• Capturing N dynamic video signals of the pedestrians and sending the respectively captured multiple dynamic video signals of the pedestrians to a central processor by N omnidirectional cameras, where N is a positive integer;
• Splicing and compressing the N dynamic pedestrian video signals and sending the video data obtained after splicing and compression to a display screen by the central processor;
• Decoding the spliced and compressed video data through the display screen to obtain decoded video data and displaying the decoded video data on the display screen to display pedestrian dynamics on the display screen.

• omnidirektionale Kameras, die zum Erfassen von N dynamischen Videosignalen der Fußgänger und zum Senden der jeweils erfassten mehreren dynamischen Videosignalen der Fußgänger an einen Zentralprozessor ausgebildet werden, wobei N eine positive ganze Zahl ist;
• einen Zentralprozessor, der zum Spleißen sowie Komprimieren der N dynamischen Videosignalen der Fußgänger und zum Senden der nach dem Spleißen sowie Komprimieren erhaltenen Videodaten an einen Anzeigebildschirm ausgebildet ist;
• einen Anzeigebildschirm, der zum Decodieren der gespleißten sowie komprimierten Videodaten ausgebildet wird, um decodierte Videodaten zu erhalten, und auf dem die decodierten Videodaten angezeigt sind, um die Fußgängerdynamik auf dem Anzeigebildschirm anzuzeigen.

To achieve the above object, the present invention further provides a system for displaying pedestrian dynamics, the system comprising:

• omnidirectional cameras, which are designed to capture N dynamic video signals of the pedestrians and to send the respectively captured multiple dynamic video signals of the pedestrians to a central processor, where N is a positive integer;
• a central processor configured to splicing and compressing the N dynamic pedestrian video signals and sending the video data obtained after splicing and compression to a display screen;
• a display screen configured to decode the spliced and compressed video data to obtain decoded video data and on which the decoded video data is displayed to display pedestrian dynamics on the display screen.

In comparison to the prior art, the present invention proposes a method for displaying the pedestrian dynamics, wherein in the present invention N omnidirectional cameras capture N dynamic video signals of the pedestrians and send the respectively captured multiple dynamic video signals of the pedestrians to a central processor, where N is a positive one is an integer, the central processor splicing and compressing the N dynamic pedestrian video signals and sending the video data obtained after splicing and compression to a display screen, the display screen decoding the spliced and compressed video data to obtain decoded video data, and on the Display screen, the decoded video data is displayed to show the pedestrian dynamics on the display screen. Thus, by watching the video on the display screen, a cyclist can know the pedestrian dynamics under the ramp and brake in good time if there are more pedestrians under the ramp, thereby ensuring traffic safety.

Figure list

• 1 zeigt ein schematisches Flussdiagramm eines Verfahrens zum Anzeigen der Fußgängerdynamik gemäß einem ersten Ausführungsbeispiel der vorliegenden Erfindung;
• 2 zeigt ein schematisches Flussdiagramm der Teilschritte von Schritt S102 aus 1;
• 3 zeigt ein schematisches Flussdiagramm zusätzlicher Schritte für 2;
• 4 zeigt ein schematisches Flussdiagramm der Teilschritte von Schritt S301 aus 3;
• 5 zeigt ein schematisches Diagramm von Funktionsmodulen eines Systems zum Anzeigen der Fußgängerdynamik gemäß einem dritten Ausführungsbeispiel der vorliegenden Erfindung;
• 6 zeigt ein schematisches Diagramm von Teilfunktionsmodulen des Zentralprozessors 502 aus 5;
• 7 zeigt ein schematisches Diagramm zusätzlicher Funktionsmodule des Zentralprozessors 502 aus 5;
• 8 zeigt ein schematisches Diagramm von Teilfunktionsmodulen des Berechnungsmoduls 701 aus 7.

In order to more clearly illustrate the exemplary embodiments of the present invention or the technical configurations in the prior art, the drawings which are used in the exemplary embodiments or the description of the prior art are briefly described below. Obviously, the drawings described below represent only a few exemplary embodiments of the present invention. Those skilled in the art can also obtain further drawings based on these drawings without performing any inventive work.

• 1 shows a schematic flow diagram of a method for displaying the pedestrian dynamics according to a first embodiment of the present invention;
• 2nd shows a schematic flow diagram of the substeps of step S102 out 1 ;
• 3rd shows a schematic flow diagram of additional steps for 2nd ;
• 4th shows a schematic flow diagram of the substeps of step S301 out 3rd ;
• 5 shows a schematic diagram of functional modules of a system for displaying the pedestrian dynamics according to a third embodiment of the present invention;
• 6 shows a schematic diagram of sub-function modules of the central processor 502 out 5 ;
• 7 shows a schematic diagram of additional function modules of the central processor 502 out 5 ;
• 8th shows a schematic diagram of sub-function modules of the calculation module 701 out 7 .

DETAILED DESCRIPTION

In order that the object, the features and the advantages of the present invention become more obvious and easier to understand, the technical configurations in exemplary embodiments of the present invention are described clearly and completely below with reference to the accompanying drawings of the exemplary embodiments of the present invention. The exemplary embodiments described are obviously only part of the exemplary embodiments of the invention and not all exemplary embodiments. All other exemplary embodiments, which the person skilled in the art can achieve on the basis of the exemplary embodiments in the present invention without creative effort, are intended to fall within the scope of protection of the present invention.

In order to explain the technical configurations described in the present invention, specific exemplary embodiments are described below.

• Schritt S101: Erfassen von N dynamischen Videosignalen der Fußgänger und Senden der jeweils erfassten mehreren dynamischen Videosignalen der Fußgänger an einen Zentralprozessor durch N omnidirektionale Kameras, wobei N eine positive ganze Zahl ist;

In 1 A schematic flow diagram of a method for displaying the pedestrian dynamics according to a first exemplary embodiment of the present invention is shown, comprising:

• step S101 : Capturing N dynamic video signals of the pedestrians and sending the respectively captured multiple dynamic video signals of the pedestrians to a central processor by N omnidirectional cameras, where N is a positive integer;

In the present exemplary embodiment of the present invention, the omnidirectional cameras are arranged under a bicycle ramp, and their positions can also be adjusted as required.

Each of the omnidirectional cameras compresses the dynamic pedestrian video signal after capturing N dynamic pedestrian video signals and sends the compressed, N dynamic pedestrian video signals to the central processor, thereby improving the efficiency of the transmission.

step S102 : Splicing and compressing the N dynamic pedestrian video signals and sending the video data obtained after splicing and compression to a display screen by the central processor;

in the present embodiment of the present invention, the central processor splices the N dynamic video signals of the pedestrians into a panorama image using a panorama connection technique and compresses the panorama image to obtain code-compressed video data; and the code-compressed video data is sent to the display screen.

step S 103 : Decoding the spliced and compressed video data through the display screen to obtain decoded video data, and displaying the decoded video data on the display screen to display pedestrian dynamics on the display screen.

In comparison with the prior art, exemplary embodiments of the present invention provide that omnidirectional cameras capture N dynamic video signals from the pedestrians and the plurality of captured in each case send pedestrian dynamic video signals to a central processor, where N is a positive integer, the central processor splicing and compressing the pedestrian N dynamic video signals, and sending the video data obtained after splicing and compression to a display screen, the display screen splicing and splicing compressed video data is decoded to obtain decoded video data, and the decoded video data is displayed on the display screen to indicate pedestrian dynamics on the display screen. A cyclist can thus know the dynamics of pedestrians under the ramp by watching the video on the display screen and can brake in good time if there are more pedestrians under the ramp, thus ensuring traffic safety.

• Schritt S201: Spleißen der N dynamischen Videosignalen der Fußgänger durch den Zentralprozessor unter Verwendung einer Panorama-Verbindungstechnik in ein Panoramabild;

In 2nd Figure 3 is a schematic flow diagram of the substeps of step S102 out 1 shown, including:

• step S201 : Splicing the N dynamic pedestrian video signals by the central processor using a panorama connection technique into a panoramic image;

In the present exemplary embodiment of the present invention, the N dynamic video signals of the pedestrians are spliced into a panorama image using a panorama connection technique, as a result of which a seamless integration of multiple images is achieved.

step S202 : Code compression of the panoramic image using image compression coding technology to produce a compressed image;

In the present exemplary embodiment of the present invention, memory space can be saved by the code compression of the panoramic image, wherein the transmission efficiency can also be improved during transmission.

step S203 : Send the compressed image to the display screen.

In the present embodiment of the present invention, the central processor splices the N dynamic video signals of the pedestrians into a panorama picture using a panorama connection technique, thereby achieving seamless integration of multiple pictures, the panorama picture being code-compressed using an image compression coding technology to produce a compressed picture , which is sent to the display screen to improve transmission efficiency, the display screen decoding the compressed image to obtain decoded video data and the decoded video data being displayed on the display screen to display the pedestrian dynamics on the display screen. Thus, by watching the video on the display screen, the cyclist can know the pedestrian dynamics, under the ramp and slow down in good time if there are more pedestrians under the ramp, which ensures traffic safety.

• Schritt S301: Berechnen einer Kopfzahl in dem Panoramabild durch den Zentralprozessor basierend auf dem Panoramabild;
• Schritt S302: Durchführen einer akustischen und visuellen Warnaktion, wenn die Kopfzahl einen voreingestellten Schwellenwert überschreitet.

In 3rd is a schematic flow diagram of additional steps for 2nd shown, especially after the step S201 out 2nd the following steps are provided:

• step S301 : Computing a head count in the panoramic image by the central processor based on the panoramic image;
• step S302 : Perform an audible and visual warning action when the head count exceeds a preset threshold.

In the present exemplary embodiment of the present invention, an acoustic device and a lighting device are arranged on the bicycle ramp. If the central processor calculates that the head count exceeds the preset threshold, there are a relatively large number of pedestrians and an acoustic and visual warning action is carried out, the central processor controlling the acoustic device for outputting a speech reminder and the lighting device for lighting in order to set the cyclists to a relatively large number Point out pedestrians in front of them and to slow them down, which ensures traffic safety.

• Schritt S401: Interframe-Differenzieren des Panoramabilds basierend auf einem Interframe-Differenzverfahren, um eine Fußgängerzone zu erhalten;
• im vorliegenden Ausführungsbeispiel der vorliegenden Erfindung wird ein aktuelles Einzelbild des Panoramabildes mit einem vorherigen Einzelbild differenziert, um ein Differenzbild zu erhalten, wonach das Differenzbild binarisiert wird, um ein Binärbild zu erhalten, in dem ein Punkt mit einem Pixelwert von 0 ein Hintergrundpunkt ist und ein Punkt mit einem Pixelwert von 1 einen Fußgänger darstellt.

In 4th . Figure 3 is a schematic flow diagram of the substeps of step S301 out 3rd shown, including:

• step S401 : Interframe differentiating the panoramic image based on an interframe difference method to obtain a pedestrian zone;
• in the present embodiment of the present invention, a current frame of the panoramic image is differentiated with a previous frame to obtain a difference image, after which the difference image is binarized to obtain a binary image in which a dot with a pixel value of 0 is a background point and on Point with a pixel value of 1 represents a pedestrian.

step S402 : Detecting the pedestrian zone with a plurality of preset human head classifiers to obtain individual head images, the plurality of preset human head classifiers being obtained by performing a plurality of training sample sets training a support vector machine (SVM) classifier;
In the present exemplary embodiment of the present invention, because of the training of the large number of training sample sets by the SVM classifiers, a large number of classifiers of the human heads are obtained, the training sample sets being divided into two types of training sample sets, namely with a positive sample set Header images and a negative sample set without header images, where all negative samples in the positive sample set and all negative samples in the negative sample set are grayscale images with 20 * 20 pixels. Depending on the human heads recorded in different orientations, the positive sample set can in turn be subdivided into five different positive partial sample sets, a positive partial sample set from above, a positive partial sample set from the front, a positive partial sample set from the back, a positive partial sample set from the left and a positive partial sample set from the right side. At the same time, the negative sample set can be divided into a negative partial sample set from above, a negative partial sample set from the front, a negative partial sample set from the back, a negative partial sample set from the left and a negative partial sample set from the right side. The SVM classifier is used to train each positive sub-sample set and negative sub-sample set so that five types of classifiers are obtained, which are a top classifier, a front classifier, a back classifier, a left classifier and one Include the right side classifier.

In the present exemplary embodiment of the present invention, an integral map calculation is carried out for the pedestrian zone in order to generate an integral map, the rows and columns of the integral map being slidably scanned in steps of 3 pixels in the horizontal and vertical directions based on a preset detection window, a hair Feature is calculated within the capture window, with the hair feature being entered into the plurality of human head classifiers, with an area where the capture window is currently located determined as a single header image when "Yes" is output.

An image block of 20 * 20 pixels is used as a preset detection window.

step S403 : Determine the number of headers in the pedestrian zone by adding all the individual header images obtained in the pedestrian zone.

In the present exemplary embodiment of the present invention, the interframe differentiation for the panoramic image is carried out based on an interframe differentiation method in order to obtain a pedestrian zone, the pedestrian zone being detected using a plurality of preset classifiers of the human heads in order to obtain individual header images, the A plurality of preset human head classifiers are obtained by training a plurality of training sample sets by an SVM classifier, the head count in the pedestrian zone being determined by adding all of the individual header images obtained in the pedestrian zone. With this algorithm, the head count can be determined quickly and accurately. If the head count exceeds the preset threshold, there are a relatively large number of pedestrians and an acoustic and visual warning action is carried out, the acoustic device emitting a speech reminder and the lighting device emitting light in order to alert the cyclists to a relatively large number of pedestrians in front of them and to slow them down, thereby road safety is guaranteed.

• N omnidirektionale Kameras, die zum Erfassen von N dynamischen Videosignalen der Fußgänger und zum Senden der jeweils erfassten mehreren dynamischen Videosignalen der Fußgänger an einen Zentralprozessor ausgebildet werden, wobei N eine positive ganze Zahl ist;
• im vorliegenden Ausführungsbeispiel der vorliegenden Erfindung sind die N omnidirektionalen Kameras 501 unter einer Fahrradrampe angeordnet, wobei ihre Positionen auch je nach Bedarf verstellt werden können.

In 5 A schematic diagram of functional modules of a system for displaying pedestrian dynamics according to a third exemplary embodiment of the present invention is shown, comprising:

• N omnidirectional cameras, which are designed to capture N dynamic video signals of the pedestrians and to send the respectively captured multiple dynamic video signals of the pedestrians to a central processor, where N is a positive integer;
• in the present exemplary embodiment of the present invention, the omnidirectional cameras 501 arranged under a bicycle ramp, whereby their positions can also be adjusted as required.

Each of the omnidirectional cameras compresses the dynamic pedestrian video signal after capturing N dynamic pedestrian video signals and sends the compressed, N dynamic pedestrian video signals to the central processor, thereby improving the efficiency of the transmission.
a central processor 502 which is designed for splicing and compressing the N dynamic video signals of the pedestrians and for sending the video data obtained after splicing and compressing to a display screen;
in the present exemplary embodiment of the present invention, the central processor splices 502 the N dynamic pedestrian video signals into a panoramic image using a panoramic connection technique and compresses the panoramic image to obtain code-compressed video data; and the code-compressed video data is sent to the display screen.
a display screen 503 which is designed to decode the spliced and compressed video data to obtain decoded video data and on which the decoded video data is displayed in order to display the pedestrian dynamics on the display screen.

In comparison to the prior art, exemplary embodiments of the present invention provide that omnidirectional cameras 501 N capture dynamic video signals of the pedestrians and the respectively captured multiple dynamic video signals of the pedestrians to a central processor 502 send, where N is a positive integer, where the central processor 502 the N dynamic video signals of the pedestrians spliced and compressed and the video data obtained after the splicing and compression on a display screen 503 sends, with the display screen 503 the spliced and compressed video data is decoded to obtain decoded video data and on the display screen 503 the decoded video data is displayed to show pedestrian dynamics on the display screen. Thus, a cyclist can watch the video on the display screen 503 Know the pedestrian dynamics under the ramp and brake in good time if there are more pedestrians under the ramp, which ensures traffic safety.

• ein Spleißmodul 601, das zum Spleißen der N dynamischen Videosignalen der Fußgänger unter Verwendung einer Panorama-Verbindungstechnik in ein Panoramabild ausgebildet ist;
• im vorliegenden Ausführungsbeispiel der vorliegenden Erfindung sind die N dynamischen Videosignale der Fußgänger unter Verwendung einer Panorama-Verbindungstechnik in ein Panoramabild gespleißt, wodurch eine nahtlose Integration mehrerer Bilder erreicht wird.
• ein Codekomprimierungmodul 602, das zum Codekomprimierung des Panoramabilds unter Verwendung einer Bildkomprimierungscodierungstechnologie ausgebildet ist, um ein komprimiertes Bild zu erzeugen;
• im vorliegenden Ausführungsbeispiel der vorliegenden Erfindung codekomprimiert das Codekomprimierungmodul 602 das Panoramabild, was den Speicherplatz sparen kann, wobei beim Übertragen die Übertragungseffizienz auch verbesserbar ist.
• ein Sendemodul 603, das zum Senden des komprimierten Bilds an den Anzeigebildschirm ausgebildet ist.

In 6 is a schematic diagram of sub-function modules of the central processor 502 out 5 shown, including:

• a splice module 601 , which is designed for splicing the N dynamic video signals of the pedestrians into a panorama image using a panorama connection technique;
• In the present exemplary embodiment of the present invention, the N dynamic video signals of the pedestrians are spliced into a panorama image using a panorama connection technique, as a result of which a seamless integration of multiple images is achieved.
• a code compression module 602 configured to code compress the panoramic image using image compression encoding technology to produce a compressed image;
• in the present embodiment of the present invention, the code compression module code-compresses 602 the panoramic image, which can save storage space, while the transmission efficiency can also be improved during transmission.
• a transmitter module 603 , which is designed to send the compressed image to the display screen.

In the present exemplary embodiment of the present invention, the splicing module splices 601 the N dynamic pedestrian video signals using a panoramic connection technique in a panoramic image, thereby achieving seamless integration of multiple images, the code compression module 602 the panoramic image is codecompressed using image compression encoding technology to produce a compressed image, the transmitter module 603 sends the compressed image to the display screen to improve transmission efficiency, the display screen decoding the compressed image to obtain decoded video data, and displaying the decoded video data on the display screen to display pedestrian dynamics on the display screen. Thus, by watching the video on the display screen, the cyclist can know the pedestrian dynamics under the ramp and brake in good time if there are more pedestrians under the ramp, thereby ensuring traffic safety.

• ein Berechnungsmodul 701, das zum Berechnen einer Kopfzahl in dem Panoramabild basierend auf dem Panoramabild ausgebildet ist;
• und ein akustisches und visuelles Warnmodul 702, das zum Durchführen einer akustischen und visuellen Warnaktion ausgebildet ist, wenn die Kopfzahl einen voreingestellten Schwellenwert überschreitet.

In 7 is a schematic diagram of additional function modules of the central processor 502 out 5 shown, including:

• a calculation module 701 configured to calculate a head count in the panoramic image based on the panoramic image;
• and an acoustic and visual warning module 702 , which is designed to carry out an acoustic and visual warning action when the head count exceeds a preset threshold.

In the present exemplary embodiment of the present invention, an acoustic device and a lighting device are arranged on the bicycle ramp. If the head count exceeds the preset threshold, there are relatively many pedestrians and the acoustic and visual warning module 702 an acoustic and visual warning action, the acoustic and visual warning module 702 control the acoustic device for outputting a speech reminder and the lighting device for illuminating, in order to alert the cyclists to a relatively large number of pedestrians in front of them and to slow them down, thereby ensuring traffic safety.

• eine Berechnungseinheit 801, die zum Interframe-Differenzieren des Panoramabilds basierend auf einem Interframe-Differenzverfahren ausgebildet ist, um eine Fußgängerzone zu erhalten;
• im vorliegenden Ausführungsbeispiel der vorliegenden Erfindung differenziert die Berechnungseinheit 801 ein aktuelles Einzelbild des Panoramabildes mit einem vorherigen Einzelbild, um ein Differenzbild zu erhalten, wonach das Differenzbild binarisiert wird, um ein Binärbild zu erhalten, in dem ein Punkt mit einem Pixelwert von 0 ein Hintergrundpunkt ist und ein Punkt mit einem Pixelwert von 1 einen Fußgänger darstellt.
• eine Erfassungseinheit 802, die zum Erfassen der Fußgängerzone mit einer Vielzahl von voreingestellten Klassifikatoren der Menschköpfe ausgebildet ist, um einzelne Kopfbilder zu erhalten, wobei die Vielzahl von voreingestellten Klassifikatoren der Menschköpfe erhalten werden, indem eine Vielzahl von Trainings-Probensätzen durch einen SVM- Klassifikator trainiert werden;
• im vorliegenden Ausführungsbeispiel der vorliegenden Erfindung werden aufgrund des Trainierens der Vielzahl von Trainings-Probensätzen durch die SVM-Klassifikatoren eine Vielzahl von Klassifikatoren der Menschköpfe erhalten, wobei die Trainings-Probensätzen in zwei Arten von Trainings-Probensätzen gegliedert werden, und zwar einem positiven Probensatz mit Kopfbildern und einem negativen Probensatz ohne Kopfbilder, wobei alle negativen Proben im positiven Probensatz und alle negativen Proben im negativen Probensatz Graustufenbilder mit 20 * 20 Pixeln sind. Abhängig von den in verschiedenen Ausrichtungen erfassten Menschköpfen kann der positive Probensatz wiederum in fünf verschiedene positive Teilprobensätze unterteilt werden, die einen positiven Teilprobensatz von oben, einen positiven Teilprobensatz aus Vorderseite, einen positiven Teilprobensatz aus Rückseite, einen positiven Teilprobensatz aus linker Seite und einen positiven Teilprobensatz aus rechter Seite umfassen. Zur gleichen Zeit kann der negative Probensatz in einen negativen Teilprobensatz von oben, einen negativen Teilprobensatz aus Vorderseite, einen negativen Teilprobensatz aus Rückseite, einen negativen Teilprobensatz aus linker Seite und einen negativen Teilprobensatz aus rechter Seite unterteilt werden. Der SVM-Klassifikator wird verwendet, um jeden positiven Teilprobensatz und jeden negativen Teilprobensatz zu trainieren, so dass fünf Arten von Klassifikatoren erhalten sind, die einen Klassifikator von oben, einen Klassifikator aus Vorderseite, einen Klassifikator aus Rückseite, einen Klassifikator aus linker Seite und einen Klassifikator aus rechter Seite umfassen.

In 8th is a schematic diagram of sub-function modules of the calculation module 701 out 7 shown, including:

• a calculation unit 801 configured to interframe differentiate the panoramic image based on an interframe difference method to obtain a pedestrian zone;
• In the present exemplary embodiment of the present invention, the calculation unit differentiates 801 a current frame of the panoramic image with a previous frame to obtain a difference image, after which the difference image is binarized to obtain a binary image in which a point with a pixel value of 0 is a background point and a point with a pixel value of 1 is a pedestrian represents.
• a registration unit 802 configured to detect the pedestrian zone with a plurality of preset human head classifiers to obtain individual head images, the plurality of preset human head classifiers being obtained by training a plurality of training sample sets by an SVM classifier;
• In the present exemplary embodiment of the present invention, because of the training of the large number of training sample sets by the SVM classifiers, a large number of classifiers of the human heads are obtained, the training sample sets being divided into two types of training sample sets, namely with a positive sample set Header images and a negative sample set without header images, where all negative samples in the positive sample set and all negative samples in the negative sample set are grayscale images with 20 * 20 pixels. Depending on the human heads recorded in different orientations, the positive sample set can in turn be subdivided into five different positive partial sample sets, a positive partial sample set from above, a positive partial sample set from the front, a positive partial sample set from the back, a positive partial sample set from the left and a positive partial sample set from the right side. At the same time, the negative sample set can be divided into a negative partial sample set from above, a negative partial sample set from the front, a negative partial sample set from the back, a negative partial sample set from the left and a negative partial sample set from the right side. The SVM classifier is used to train each positive sub-sample set and negative sub-sample set so that five types of classifiers are obtained, which are a top classifier, a front classifier, a back classifier, a left classifier and one Include the right side classifier.

In the present exemplary embodiment of the present invention, an integral map calculation is carried out for the pedestrian zone in order to generate an integral map, the rows and columns of the integral map being slidably scanned in steps of 3 pixels in the horizontal and vertical directions based on a preset detection window, a hair Feature is calculated within the capture window, with the hair feature being entered into the plurality of human head classifiers, with an area where the capture window is currently located determined as a single header image when "Yes" is output.

An image block of 20 * 20 pixels is used as a preset detection window.
and a determination unit 803 which are used to determine the number of headers in the pedestrian zone by adding up all the individual header images obtained in the pedestrian zone.

In the present exemplary embodiment of the present invention, the calculation unit performs 801 performing the interframe differentiation for the panoramic image based on an interframe difference method to obtain a pedestrian zone, the detection unit 802 the pedestrian zone is captured with a plurality of preset human head classifiers to obtain individual head images, the plurality of preset human head classifiers being obtained by training a plurality of training sample sets by an SVM classifier, the determination unit 803 the number of headers in the pedestrian zone is determined by adding all the individual header images obtained in the pedestrian zone. With this algorithm, the head count can be determined quickly and accurately. If the head count exceeds the preset threshold, there are a relatively large number of pedestrians and an acoustic and visual warning action is carried out, the acoustic device emitting a speech reminder and the lighting device emitting light in order to alert the cyclists to a relatively large number of pedestrians in front of them and to slow them down, thereby road safety is guaranteed.

In the various embodiments provided by the present application, it is to be understood that the disclosed apparatus and method may be made possible in other ways. For example, the device embodiments described above are for illustration only. For example, the division of the modules represents only a logical division of functions. For example, the division of the module is only a logical division of functions, while further distribution methods can be used in the actual implementation, for example several modules or components can be combined with one another or integrated into another system, or some features can be omitted or not performed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may not only be an indirect coupling or communication connection via some interfaces, devices or modules, but also an electrical, mechanical connection and the like.

The modules described as separate components may or may not be physically separate. The components displayed as modules may or may not be physical modules, that is to say they can be located in one place or distributed over several network modules. Some or all of the modules below may be selected according to actual needs to achieve the purpose of the embodiment of the present embodiment.

In addition, all of the functional modules in the single embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules.

If the integrated modules are implemented in the form of software function modules and are sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the part of the technical embodiment of the present invention that is essential or contributes to the prior art, or all or part of this technical embodiment can be implemented in the form of a software product that is stored on a storage medium and contains some instructions to cause a computing device, which may be a personal computer, a server or a network device, etc., to perform all or part of the steps of the method in various embodiments of the present invention. The above-mentioned storage medium contains various media on which program code can be stored and which, for example, a USB stick, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM, random access memory), a magnetic disk or a Is CD and the like.

It should be noted that, for simplicity and brevity, the above various embodiments of the method are described as a series of combinations of actions. However, those skilled in the art should understand that the present invention is not limited to the order of actions described, since according to the present invention some steps can be carried out in a different order or simultaneously. Second, those skilled in the art should also understand that the embodiments described in the description are preferred embodiments and that the actions and modules involved are not necessarily required by the present invention.

Different exemplary embodiments are described here with their own different focal points. A part that is not explained in detail in one exemplary embodiment can be referred to the associated specification of other exemplary embodiments.

The above describes a method and system for displaying pedestrian dynamics. One skilled in the art can change the specific embodiments and the scope according to the concept of the embodiments of the present invention. In summary, the content of the description should not be construed as limiting the invention.

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• CN 201710376818 [0001]

Claims (5)

A system for displaying pedestrian dynamics, comprising: omnidirectional cameras configured to capture N dynamic video signals from the pedestrians and to transmit the respective captured multiple dynamic video signals from the pedestrians to a central processor, where N is a positive integer; a central processor configured to splicing and compressing the N dynamic pedestrian video signals and sending the video data obtained after splicing and compression to a display screen; a display screen configured to decode the spliced and compressed video data to obtain decoded video data and on which the decoded video data is displayed to display pedestrian dynamics on the display screen. System according to Claim 1 , characterized in that the central processor comprises: a splicing module which is designed for splicing the N dynamic video signals of the pedestrians into a panorama image using a panorama connection technique; a code compression module configured to code compress the panoramic image using image compression coding technology to produce a compressed image; and a transmission module configured to transmit the compressed image to the display screen. System according to Claim 2 , characterized in that the system further comprises: a calculation module configured to calculate a head count in the panoramic image based on the panoramic image; and an audible and visual warning module configured to perform an audible and visual warning action when the head count exceeds a preset threshold. System according to Claim 3 , characterized in that the computation module comprises: a computation unit configured to interframe differentiate the panoramic image based on an interframe difference method to obtain a pedestrian zone; a detection unit configured to detect the pedestrian zone with a plurality of preset human head classifiers to obtain individual head images, the plurality of preset human head classifiers being obtained by training a plurality of training sample sets by a classifier of the support vector machine ; and a determination unit for determining the number of headers in the pedestrian zone by adding all the individual header images obtained in the pedestrian zone. System according to Claim 4 , characterized in that the detection unit is specifically designed: for integral image calculation of the pedestrian zone in order to generate an integral map; for slidingly scanning the integral card based on a preset acquisition window and for calculating a hair feature in the acquisition window; for entering the hair feature in the plurality of classifiers of the human heads, and for determining an area in which the detection window is currently located as a single header image if "Yes" is output.

Images