CN217562054U - Car based on posture gesture recognition shoots - Google Patents

Abstract

The utility model discloses a car based on posture gesture recognition shoots, relate to intelligent vehicle control, accuse screen or cell-phone end trigger the car in the car and shoot the order and send to accuse module in the car, automatic gearbox control module feeds back current gear and speed of a motor vehicle to well accuse module, well accuse module starts to open the camera and carries out comparison discernment with the posture gesture of prestoring to acquireing image transmission, and to well accuse module feedback identification result, if the posture gesture of target image and prestoring is unanimous, well accuse module sends the car light control order and triggers body controller control car light flicker, trigger the camera and start after the car light flicker predetermined length and shoot, the photo storage of shooting to car machine storage module. The utility model discloses can extensively be used for intelligent car.

Description

Car based on posture gesture recognition shoots

Technical Field

The utility model relates to an intelligent automobile technical field specifically is the outside scenery of utilizing the camera of vehicle self to shoot, personage etc. automatically.

Background

The current posture recognition technology which takes human body, posture and action as characteristics starts to break through continuously, the accuracy rate is improved continuously, and a huge application prospect is shown. In the automotive field, the way human-vehicle interaction is constantly changing, and in recent years this trend has become more and more apparent. In the vehicle infotainment system only, the interaction mode starts to change from the earliest physical key pressing to a multi-dimensional interaction mode comprising key pressing, touch screen, voice and the like. Gesture recognition control and posture recognition control are becoming a more desirable in-vehicle interaction mode, for example, through different gestures, the gesture interaction can realize functions of hanging up a phone, adjusting volume, selecting songs, controlling navigation, controlling vehicles (air conditioners, seats, windows and the like), and the like, and also comprises various scenes of interaction of a main driver, a secondary driver and passengers in a back row.

However, the exploration of the application scene outside the vehicle is very little. Such as publication No.: CN206568977U, entitled "vehicle mounted gesture control device and vehicle" in the patent of the utility model discloses a chinese utility model mentioned that at least one camera is utilized to collect gesture data of a driver from a plurality of angles, only recognizes user gestures in the vehicle, but does not make explicit use of the recognition and application outside the vehicle.

SUMMERY OF THE UTILITY MODEL

The utility model discloses develop the function of shooing outside the car based on intelligent automobile, go deep into the scene of shooing outside the car when going out a journey, through the ability that the car machine camera caught the image, the automatic recognition picture, combine the feedback of car machine light, loudspeaker, realize that the user can conveniently shoot outside the car using the car machine camera; through the data transmission capability of the local area network, a user can quickly browse photos shot by the vehicle machine outside the vehicle.

The utility model provides an above-mentioned technical problem's technical scheme be, an automobile control equipment of shooing based on posture gesture recognition, include: the system comprises a vehicle control module, an automatic gearbox control module, a camera module, a processor, a vehicle body controller and a vehicle storage module, wherein the camera module is arranged at the front end of a vehicle, a lens faces the front of the vehicle head, the vehicle control screen and the automatic gearbox control module are connected with the input end of the central control module through a vehicle body bus, the central control module receives a vehicle exterior shooting command and the current gear and vehicle speed, the output end of the central control module is connected with the camera module and the vehicle body controller, the camera module dynamically acquires images and outputs the images to the processor to be compared and identified with prestored body posture gestures, the processor feeds identification results back to the central control module, if a target image body posture gesture is consistent with the prestored body posture gesture, the central control module outputs the command to trigger the camera module to start shooting, and the camera acquires the images and outputs the acquired images to the vehicle storage module to be stored.

Furthermore, the camera module is arranged on the rearview mirror of the automobile or at the position where the edge of the front windshield does not shield the sight of the driver. If the 800-1000w pixel camera module is arranged on the automobile rearview mirror, the human body is positioned in the range of 2.5m-10m away from the right front of the camera when the posture gesture is stored in advance.

Furthermore, the vehicle-mounted storage module establishes a hot spot connection with the mobile phone WLAN through the vehicle-mounted WLAN. The mobile phone sends a photographing command outside the vehicle to the vehicle-mounted central control module, and the pictures stored in the vehicle-mounted storage module are pushed to the mobile phone end for preview and downloading.

The vehicle control module sends a vehicle lamp control command to trigger the vehicle body controller to control the vehicle lamp to flicker, and the camera is started to take a picture after the vehicle lamp flickers for a preset time. The pre-stored posture gesture is that the left-hand lifting angle of a standing human body in a preset range is larger than 120 degrees, or the two-hand lifting angle is larger than 120 degrees, or the right-hand lifting angle is larger than 120 degrees. When more than 3 frames of photos in the current captured target image 1-2s video stream segment are consistent with the pre-stored body gesture, the vehicle central control module triggers a horn to sound for a short time to the vehicle body controller, triggers the shooting countdown, and controls the camera to delay the preset time for shooting.

The utility model discloses based on the relevant hardware facilities of intelligent automobile, can be based on the function of shooing of attitude automatic start car, solve convenience, the promptness that the user took a picture outside the car, the car group is taken a photograph and is taken a photograph with the scenery jointly to can browse fast outside the car and use the photo that the car machine was shot.

Drawings

FIG. 1 is a schematic diagram of the operation of a vehicle exterior posture control photographing control module;

FIG. 2 is a schematic diagram of a pre-stored posture standing position area;

fig. 3 is a layout diagram of a camera module, a central control module and the like.

Detailed Description

The following detailed description of the embodiments of the invention is provided in connection with the accompanying drawings.

A user utilizes a vehicle camera at a rearview mirror to photograph people or scenery outside the vehicle, and the vehicle camera module acquires and sends the certain gesture or posture to a processor for recognition by swinging a certain gesture or posture in front of the vehicle, and then the photographing is automatically triggered; also can be through the cell-phone of connecting the car machine, through starting cell-phone app application, take a picture at the outer remote control car machine of car.

The car body state photographing mainly comprises a) automatically photographing pictures after recognizing the gesture of a user; b) The mobile phone is connected with the car machine through the WIFI to transmit the pictures to the mobile phone end, so that the user can preview the currently shot pictures in real time when the car-exterior body photographing is carried out.

The body state control is shot, the in-vehicle shooting function unit is triggered to start the outside-vehicle body state shooting mode, the in-vehicle control module triggers safety check, a safety check command is sent to the automatic gearbox control module, and the automatic gearbox control module feeds back the current gear and speed.

The method comprises the steps of opening a vehicle-mounted photographing APP, selecting a posture photographing mode, judging whether a vehicle end is a P gear currently or not, starting a WIFI hotspot, connecting a vehicle WIFI with a mobile phone, capturing a target in a preset range by a vehicle end camera module, triggering a turn light to flicker for a preset time after a processor identifies that the target implements a specified posture action, starting the camera module to photograph when the light is on for a preset time each time, storing a photo in a vehicle memory, transmitting the photo thumbnail to the mobile phone through the WIFI to synchronously view, and selecting and determining the photo to be stored in the mobile phone memory.

Fig. 1 is a working schematic diagram of a vehicle exterior body state control photographing control module, wherein a vehicle interior control screen (which can also be triggered by a mobile phone terminal) triggers a vehicle exterior photographing command to be sent to the vehicle interior control module, an automatic transmission control module feeds back a current gear and speed of a vehicle to a central control module, the central control module starts to turn on a camera to dynamically capture an image within a preset range, the obtained image is transmitted to a processor to be compared with a preset gesture body state, and feeds back a recognition result to the central control module, if the recognition result is matched with a prestored body state gesture, the central control module triggers a vehicle lamp control command, a vehicle body controller controls vehicle lamp light to flicker, and a camera module starts to photograph after the vehicle lamp flickers for a preset time; the shot pictures are stored in the vehicle-mounted storage module, the vehicle-mounted WLAN and the mobile phone WLAN are connected through hot spots, the pictures shot by the vehicle body camera and stored in the vehicle-mounted storage module are pushed to the mobile phone end for preview, and downloading is determined.

When the current state of the vehicle meets the safety condition (such as the vehicle is in a P gear), the in-vehicle photographing function unit sends a command for starting image recognition to the camera. The camera starts to capture pictures outside the vehicle and transmits the images to the processor module. The processor module analyzes the images in the video stream, identifies the character images and key points of human body shapes, and compares the key points with gestures or body shapes preset in the system.

Identifiable gestures and human body posture can be defined according to the capability and the requirement of the camera. For example, with 200 ten thousand pixels, the recognizable gesture/posture is set as: the left-hand lifting angle of the human body standing in front is larger than 120 degrees, the two-hand lifting angle is larger than 120 degrees, and the right-hand lifting angle is larger than 120 degrees.

Fig. 2 is a diagram of suggesting a recommended target position when the posture is prestored according to the angle of the picture shot by the camera, so that the camera can conveniently capture the body posture and the gesture of the human body. The standing position distance is determined according to the height of the camera when the body posture and the gesture of the human body are captured. For a camera which is installed at the position of the rearview mirror in the automobile and has the height of 1.6 meters from the ground, the camera stands 2.5 meters in front of the camera.

Fig. 3 is a schematic layout diagram of one of the camera module, the central control module and the like, wherein the camera module is arranged on the automobile rearview mirror, and the lens faces the front of the automobile head. The central control module, the processor, the memory module and the gearbox control module are positioned on a central control platform below a central control screen of the vehicle machine.

When the camera captures a target in the recognition range, the gesture of the target is the same as the gesture of the preset body state recorded in the memory, and the gesture and the body state are kept for a preset time (for example, when the human body outside the vehicle is recognized to lift an arm, more than 3 pictures in the current 1-2s video stream segment meet the gesture set by the system),

the control module in the car machine triggers the order suggestion of controlling light scintillation (still can trigger loudspeaker to ring) to the automobile body controller, sends loudspeaker brief ring, car light scintillation control command, through the car light scintillation and last predetermined duration (like 3 seconds), the short ring one sound of outer loudspeaker of car indicates to have triggered the back down-time of shooing, and the functional unit of shooing sends the instruction of shooing, controls the camera and postpones predetermined duration and shoot.

When the corresponding gesture or posture is not recognized after the time (such as 10 minutes) is exceeded, the photographing function is automatically turned off, meanwhile, the processor does not send information to the central control module, and the camera module does not respond. The storage space of 'shooting outside the vehicle' is preset in the vehicle-mounted storage module, and after shooting is finished, pictures are automatically stored in a 'shooting outside the vehicle' catalog of the vehicle-mounted storage module.

The photos are transmitted to a mobile phone end, a tcp connection is established by utilizing wifi through an Android car machine and an Android mobile phone, the photos are mutually transmitted through an interface socket for calling a network protocol to communicate through an application program, the socket is an interface for interaction of the application program and a network protocol stack, and a mechanism for transmitting data packets to an application layer process is provided. The mobile phone is a client, and the client can see the picture thumbnails in the designated folder of the server and select to download to the mobile phone.

When the vehicle body state shooting is started, the vehicle end automatically starts the WALN, establishes a hot information source, and displays a hot name and a hot password. And opening the WIFI connection of the mobile phone end, connecting the WIFI hotspot of the vehicle machine, and establishing a local area network transmission mechanism. After shooting is finished, opening a corresponding local area network transmission plate block in the APP of the mobile phone terminal, automatically acquiring a thumbnail of a shot picture in the automobile through an established local area network transmission mechanism, and realizing flow-free downloading and saving to the local.

Claims (4)

1. The utility model provides an automobile based on posture gesture recognition shoots which characterized in that includes: the system comprises a vehicle-mounted central control module, an automatic gearbox control module, a camera module, a processor, a vehicle body controller and a vehicle-mounted storage module, wherein the camera module is arranged at the front end of a vehicle, and a lens faces the front of a vehicle head.

2. The car of claim 1, wherein the camera module is installed on the rear-view mirror of the car or at a position where the edge of the front windshield does not obstruct the view of the driver.

3. The automobile capable of taking pictures based on posture gesture recognition as claimed in claim 1, wherein the central control module, the processor, the memory module and the gearbox control module are located in a central control console below a central control screen of the automobile.

4. The automobile capable of photographing based on the posture gesture recognition as claimed in one of claims 1 to 3, wherein the car machine storage module further establishes a hot spot connection with a mobile phone WLAN through a car machine WLAN.

Images