Reminding method and device for preventing articles in vehicle from being lost

Abstract

The invention discloses a reminding method and device for preventing articles in a vehicle from being lost, and belongs to the technical field of vehicles. The in-vehicle article anti-loss reminding method comprises the following steps: extracting a first image feature of a current frame image collected by an in-vehicle camera; the first image feature of the current frame image is compared with a historical feature identifier marked by a historical image collected by an in-vehicle camera, and the historical feature identifier indicates a second image feature included in the historical image; analyzing whether articles are left in the vehicle according to a comparison result; and under the condition that the articles are left in the vehicle, prompt information indicating that the articles are left is sent to the information receiving end so as to remind passengers. According to the method, passengers or drivers can be reminded of articles left in the vehicle in time, and the riding experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of vehicle technology, and in particular to a method and device for reminding vehicles to prevent items from being left behind. Background Art

[0002] When driving or riding in a vehicle, the driver or passengers often leave items in the vehicle. In particular, it is difficult for passengers to retrieve the items left behind in a taxi, online car-hailing vehicle, etc. In addition, it is also difficult to find small items such as keys and mobile phones left in the vehicle without realizing it. Therefore, there is still a lot of room for improvement in the user's riding experience in the scenario of items being left behind in the car. Summary of the invention

[0003] In view of this, the present invention provides a method and device for reminding passengers and drivers of items left behind in the vehicle, which can remind passengers and drivers of items left behind in the vehicle and improve the riding experience.

[0004] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0005] In a first aspect, the present invention provides a method for preventing items from being left behind in a vehicle, comprising:

[0006] Extracting the first image feature of the current frame image collected in real time by the in-vehicle camera;

[0007] Comparing the first image feature of the current frame image with the historical feature identifier marked in the historical image collected by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image;

[0008] According to the comparison results, analyze whether there are any items left in the car;

[0009] In the case of items left behind in the vehicle, a prompt message indicating the left-behind item is sent to the information receiving end to remind the occupants.

[0010] Optionally, the vehicle article anti-lost reminder method further includes: marking a target feature identifier for the first image feature of the current frame image;

[0011] The comparing the first image feature of the current frame image with the historical feature identifier of the historical image captured by the in-vehicle camera includes:

[0012] The target feature identifier of the current frame image mark is compared with the historical feature identifier of the historical image mark.

[0013] Optionally, the comparing the first image feature of the current frame image with a historical feature identifier marked by a historical image captured by the in-vehicle camera includes:

[0014] Compare the first image feature of the current frame image with the historical feature identifier marked by the previous frame image captured by the in-vehicle camera and the identifier change status corresponding to the historical feature identifier, wherein the historical feature identifier marked by the previous frame image iteratively inherits the historical feature identifiers of the previous multiple frames of images and the identifier change status corresponding to the historical feature identifier;

[0015] According to the comparison result, a target feature identifier and an identifier change status corresponding to the target feature identifier are marked for the first image feature of the current frame image.

[0016] Optionally, marking a target feature identifier for the first image feature of the current frame image includes:

[0017] When the comparison result indicates that the real-time image contains one or more newly added image features,

[0018] For each of the newly added image features, the newly added image feature is identified as a living object or a non-living object, and a target feature identifier indicating a living object or a non-living object is marked on the newly added image feature.

[0019] Optionally, the identifying the newly added image feature as a living object or a non-living object includes:

[0020] Filtering target data corresponding to the newly added image features from data collected by an infrared camera installed in the vehicle;

[0021] According to the target data, it is determined that the newly added image feature is a living object or a non-living object.

[0022] Optionally,

[0023] The analyzing whether there is any item left in the vehicle includes:

[0024] When the comparison result indicates that the real-time image includes a newly added image feature indicating a non-living object and the target image feature indicating a living object in the real-time image disappears, the object left behind in the vehicle is determined and the newly added image feature is marked with a newly added feature identifier indicating a non-living object.

[0025] Optionally, the analyzing whether an item is left behind in the vehicle further includes:

[0026] The first historical image in which the newly added image feature appears is located, and when the first historical image contains the target image feature, the item left behind in the vehicle is determined.

[0027] Optionally, the analyzing whether an item is left behind in the vehicle further includes:

[0028] Determine a change status of a mark corresponding to the newly added image feature and a change status of a mark corresponding to the target image feature;

[0029] When the identification change status corresponding to the newly added image feature is consistent with the identification change status corresponding to the target image feature or when the start time indicated by the identification change status corresponding to the newly added image feature is after the start time indicated by the identification change status corresponding to the target image feature, the left-behind item in the vehicle is determined.

[0030] Optionally, the analyzing whether an item is left behind in the vehicle includes:

[0031] When the comparison result indicates that the target image feature indicating the biological object in the real-time image disappears, the real-time image is analyzed to determine whether there is a new image feature indicating a non-biological object that appears simultaneously with the target image feature. If so, it is determined that the object is left behind in the vehicle.

[0032] Optionally, the analyzing whether there is a newly added image feature indicating a non-living object that appears simultaneously with the target image feature in the real-time image includes:

[0033] The real-time image is compared with the target image in which the target image feature first appears, to determine whether the real-time image has a newly added image feature indicating a non-living object that appears simultaneously with the disappeared target image feature.

[0034] Optionally, the analyzing whether an item is left behind in the vehicle includes:

[0035] In the case where the comparison result indicates that there is no image feature indicative of a living organism in the real-time image,

[0036] Compare the first frame image after the door corresponding to the driver's seat of the vehicle is locked with the first frame image after the door corresponding to the driver's seat of the vehicle was locked the last time, to determine whether the real-time image has image features of non-biological objects; or directly iteratively inherit the historical feature identifier of the previous frame image corresponding to the first frame image after the door corresponding to the driver's seat is locked, to determine whether the image features of non-biological objects exist;

[0037] Identify items left in a vehicle in the presence of non-biological image features.

[0038] Optionally, the vehicle object anti-lost reminder method further includes: matching a target reminder mode for the vehicle object from among the preset multiple reminder modes according to the comparison result and the trigger conditions corresponding to the multiple reminder modes;

[0039] The sending of the prompt information indicating the lost item to the information receiving end includes: sending the prompt information indicating the lost item to the information receiving end through the target reminder mode.

[0040] In a second aspect, an embodiment of the present invention provides a device for preventing items from being left behind in a vehicle, comprising: a feature extraction module, a processing module and an interaction module, wherein:

[0041] The feature extraction module is used to extract the first image feature of the current frame image captured by the in-vehicle camera;

[0042] The processing module is used to compare the first image feature of the current frame image with the historical feature identifier marked by the historical image collected by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image; and analyze whether an item is left in the vehicle based on the comparison result;

[0043] The interaction module is used to send a prompt message indicating the left-behind item to the information receiving end to remind the occupants when the item is left behind in the vehicle.

[0044] In a third aspect, an embodiment of the present invention provides an electronic device, including:

[0045] one or more processors;

[0046] a storage device for storing one or more programs,

[0047] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for preventing items from being left behind in a vehicle as provided in the first aspect embodiment above.

[0048] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium on which a computer program for implementing a method for preventing items from being left behind in a vehicle is stored.

[0049] When the computer program is executed by the vehicle-mounted processor, the method for preventing items from being left behind in the vehicle provided in the first aspect embodiment described above is implemented.

[0050] The technical solution of the above invention has the following advantages or beneficial effects:

[0051] The technical solution provided by the embodiment of the present invention extracts the first image feature of the current frame image collected in real time by the in-vehicle camera, and compares the first image feature of the current frame image with the historical feature identifier marked by the historical image collected by the in-vehicle camera. Since the historical feature identifier indicates the second image feature included in the historical image, the difference between the first image feature and the historical feature identifier can be quickly obtained by comparing the first image feature with the historical feature identifier, so as to quickly analyze whether an item is left behind from the comparison result. In the case of an item left behind in the car, a prompt message indicating the left behind is sent to the information receiving end to remind the occupants. Therefore, the technical solution provided by the present invention can remind the occupants or the driver of left behind items in the car and improve the riding experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0053] Figure 2 1 is a schematic diagram of the main process of the method for preventing items from being left behind in a vehicle provided according to an embodiment of the present invention;

[0054] Figure 3 is a schematic diagram of the structure of the distribution of in-vehicle cameras in a vehicle according to an embodiment of the present invention;

[0055] Figure 4 is a schematic diagram of items left behind in a vehicle for the first scenario provided according to an embodiment of the present invention;

[0056] Figure 5 is a schematic diagram of items left behind in a vehicle for the second scenario according to an embodiment of the present invention;

[0057] Figure 6 is a schematic diagram of items left behind in a vehicle for the third scenario provided according to an embodiment of the present invention;

[0058] Figure 7 is a schematic diagram of items left behind in a vehicle for the fourth scenario provided according to an embodiment of the present invention;

[0059] Figure 8 1 is a schematic diagram of the main process of the method for preventing items from being left behind in a vehicle for the first scenario provided in accordance with an embodiment of the present invention;

[0060] Fig. 9 1 is a schematic diagram of the main process of the method for preventing items from being left behind in a vehicle for the second scenario provided in accordance with an embodiment of the present invention;

[0061] Fig.10 2 is a schematic diagram of the structure of a device for preventing items from being left behind in a vehicle according to an embodiment of the present invention;

[0062] Fig.11 2 is a schematic diagram of the structure of a vehicle article anti-lost reminder system according to an embodiment of the present invention;

[0063] Fig.12 It is a structural diagram of a computer system suitable for implementing the in-vehicle article anti-lost reminder according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0065] It should be pointed out that the embodiments of the present invention and the technical features therein may be combined with each other without conflict.

[0066] In addition, the terms “first” and “second” etc. involved in the embodiments of the present invention are used to distinguish features of different scopes or purposes, and are not intended to limit the number or order of features.

[0067] The vehicle involved in the embodiment of the present invention may be an internal combustion engine vehicle using an engine as a power source, a hybrid vehicle using an engine and an electric motor as a power source, an electric vehicle using an electric motor as a power source, or the like.

[0068] The following describes a technical scenario to which the technical solution provided by the embodiment of the present invention is applicable based on the system architecture on which the technical solution provided by the embodiment of the present invention relies.

[0069] Figure 1 An exemplary system architecture 100 is shown, to which an in-vehicle article anti-leftover reminder method or an in-vehicle article anti-leftover reminder device according to an embodiment of the present invention may be applied.

[0070] like Figure 1 As shown, the system architecture 100 may include a vehicle 101, a server 102, a terminal device 103 and a communication module 104. The communication module 104 is used to provide a communication medium between the vehicle 101 and the server 102, and between the server 102 and the terminal device 103. The communication module 104 may include networks of various connection types, such as wired and wireless communications such as Bluetooth, infrared communication, and data line communication, and may also include a communication base station.

[0071] Among them, vehicle 101 generally has multiple infrared cameras (not shown in the figure) installed on the roof ceiling, through which the infrared camera collects images in the cabin in real time. The ECU (Electronic Control Unit) of vehicle 101 obtains the image collected by the infrared camera in real time, and extracts a first image feature from the image, compares the first image feature with the second image feature historical feature identifier included in the historical image marked by the historical image collected by the infrared camera, and analyzes whether there are any items left behind by the passengers or the driver in the car based on the comparison result. When it is analyzed that there are left-behind items in the car, the ECU of vehicle 101 can send prompt information indicating the left-behind items to the HUD (Head Up Display) of the vehicle itself, such as the HUD on the front windshield, the HUD embedded in the door, etc. The ECU of vehicle 101 can also generate a sound signal, which triggers the sound device of vehicle 101 through the sound signal, so that a sound prompting the left-behind items is emitted through the sound device. Furthermore, for a taxi or online-hailing car ordered by a passenger through a taxi-hailing software of a terminal device 103, the vehicle 101 sends a prompt message indicating the left-behind item to the server 102, which sends the prompt message indicating the left-behind item to the terminal device 103, and prompts the passenger to take the left-behind item with him through the terminal device 103.

[0072] It should be noted that the in-vehicle item anti-lost reminder method provided in the embodiment of the present invention can be implemented by the ECU of the vehicle 101 , and accordingly, each module of the in-vehicle item anti-lost reminder device can be arranged in the ECU of the vehicle 101 .

[0073] It should be understood that Figure 1 The number of vehicles, servers and terminal devices in the embodiment is only for illustration. Any number of vehicles, servers and terminal devices may be provided as required.

[0074] Figure 2 is a schematic diagram of the main steps of a method for preventing items from being left behind in a vehicle according to an embodiment of the present invention. Figure 2 As shown, the method for preventing items from being left behind in a vehicle mainly includes the following steps:

[0075] Step S201: Extraction processing, extracting the first image feature of the current frame image captured in real time by the in-vehicle camera.

[0076] There may be more than one camera 30 in the vehicle. Figure 3The relative position relationship between the vehicle and the in-car camera 30 shown in the figure, the multiple in-car cameras 30 are arranged on the cabin ceiling, and the multiple in-car cameras 30 are arranged according to the seat distribution, that is, an in-car camera 30 is arranged above the driver's seat, above the co-pilot seat, and each passenger seat in the back row, to avoid shooting blind spots in each passenger seat, thereby ensuring the accuracy of subsequent analysis of whether items are left behind. Preferably, the in-car camera 30 is an infrared camera, which can clearly capture images in the cabin even in a relatively dark environment. In addition, the infrared camera can obtain relatively obvious differentiated shooting data for non-living objects such as mobile phones, key chains, packages, file bags, folders, etc., and living objects (the living objects generally refer to passengers). Therefore, based on the data captured by the infrared camera, it is convenient to distinguish living objects and non-living objects from the image.

[0077] The current frame image involved in this step generally refers to the latest image captured in real time by the in-vehicle camera.

[0078] The first image feature involved in this step generally refers to the image feature extracted from the current frame image, which can indicate the key features of the objects and / or passengers contained in the image, such as the type of the object, the location of the object in the cabin, whether there are passengers, the location of the passengers in the cabin, whether the key features exist in the previous frame image, etc.

[0079] After this step, the method may further include: marking a target feature identifier for the first image feature of the current frame image. Based on the target feature identifier, in the following step S202, the target feature identifier is compared with the historical feature identifier, which can improve the comparison efficiency and further improve the efficiency of analyzing whether an item is left in the vehicle.

[0080] Step S202: comparison processing, comparing the first image feature of the current frame image with the historical feature identifier marked by the historical image collected by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image.

[0081] The historical images captured by the in-vehicle camera generally refer to one or more frames of images captured before the current frame image. It can be the previous frame image corresponding to the current frame image, or other images at a certain time from the current frame image, or the first image after a specific operation (such as the first image after the door is locked, the first image after the door is opened, the first image after the vehicle is started, etc.). In addition, the historical images can also be multiple images captured by the in-vehicle camera within a past period of time.

[0082] Among them, the "first" and "second" in the first image feature and the second image feature are used to distinguish that the image features come from different types of images. Specifically, the first image feature comes from the current frame image, and the second image feature comes from the image features of the historical image. When the current frame image becomes the historical image, the first image feature of the current frame image becomes the second image feature.

[0083] Among them, the historical feature identifier is used to mark the position of each second image feature in the cabin, whether the second image feature is a biological object or a non-biological object, and in the case where the image feature is a non-biological object, the type of non-biological object, the time when the second image feature first appeared / the historical image serial number when the second image feature first appeared, whether the second image feature is a newly added image feature (i.e., it appears for the first time in the historical image), etc.

[0084] Step S203: Lost item analysis: based on the comparison result, analyze whether there is any item left in the car; if there is any item left in the car, execute step S204; if there is no item left in the car, directly end the current process.

[0085] When the comparison result indicates that the first image feature cannot completely match the historical feature identifier, and the first image feature that cannot match the historical feature identifier appears in the current frame image, it is determined that a new object has appeared in the cabin, and the new object may be accidentally left behind by a passenger.

[0086] Step S204: Reminder processing, sending a reminder message indicating the left-behind item to the information receiving end to remind the occupants.

[0087] This step may send a prompt message when it is analyzed in the above step S203 that there is a left-behind item in the car, or may send a prompt message indicating the left-behind item to the information receiving end only under certain triggering conditions such as the door being opened.

[0088] The information receiving end may be the HUD on the front windshield of the vehicle, the HUD on the window, the terminal device used by passengers to order online ride-hailing vehicles, the driver's terminal device, the vehicle display screen, the vehicle horn, etc. The prompt information can be used to remind passengers or drivers to find items left behind in the vehicle in a timely manner.

[0089] against Figure 2The provided technical solution extracts the first image feature of the current frame image captured by the in-vehicle camera in real time, and compares the first image feature of the current frame image with the historical feature identifier marked by the historical image captured by the in-vehicle camera. Since the historical feature identifier indicates the second image feature included in the historical image, the difference between the first image feature and the historical feature identifier can be quickly obtained by comparing the first image feature with the historical feature identifier, so as to quickly analyze whether an item is left behind from the comparison result. In the case of an item left behind in the car, a prompt message indicating the left behind is sent to the information receiving end to remind the occupants. Therefore, the technical solution provided by the present invention can remind the occupants or the driver of left behind items in the car and improve the riding experience.

[0090] There are two ways to implement the above step S202.

[0091] Specifically, based on marking the target feature identifier for the first image feature of the current frame image, the first implementation of step S202 may include: comparing the target feature identifier marked in the current frame image with the historical feature identifier marked in the historical image, which can improve the comparison efficiency, so as to further improve the efficiency of analyzing whether items are left behind in the vehicle.

[0092] The second implementation of step S202 may include: comparing the first image feature of the current frame image with the historical feature identifier marked by the previous frame image captured by the in-vehicle camera and the identifier change status corresponding to the historical feature identifier, wherein the historical feature identifier marked by the previous frame image iteratively inherits the historical feature identifiers of the multiple frames of images before it and the identifier change status corresponding to the historical feature identifier; according to the comparison result, marking the target feature identifier and the identifier change status corresponding to the target feature identifier for the first image feature of the current frame image. The historical feature identifier marked by the previous frame image iteratively inherits the historical feature identifiers of the multiple frames of images before it, that is, each frame image iteratively inherits the historical feature identifier of its previous frame image, so that the historical feature identifier can conveniently generate a historical feature identifier for each frame image. It is worth noting that the iterative inheritance generally means that when the previous frame image has the same image feature as the image before it, the previous frame image will inherit the historical feature identifier of the same image feature. If a certain image feature disappears in the previous frame image, the previous frame image will no longer mark the disappeared image feature.

[0093] The identification change status mainly marks the time when the historical feature identification first appears or the image sequence number of the first appearance of the historical feature identification and other historical feature identifications that appear together with the historical feature identification.

[0094] The embodiment of the present invention compares the first image feature of the current frame image or the target feature identifier corresponding to the first image feature with the historical feature identifier marked in the previous frame image and the identifier change status corresponding to the historical feature identifier, so that the comparison process has relatively small resource overhead and the efficiency of analyzing whether items are lost is improved.

[0095] More specifically, in the second implementation of step S202, the specific implementation method of marking the target feature identifier for the first image feature of the current frame image may include: when the comparison result indicates that the real-time image contains one or more newly added image features, for each newly added image feature, identify the newly added image feature as a living organism or a non-living organism, and mark the newly added image feature with a target feature identifier indicating a living organism or a non-living organism. In addition to indicating a living organism or a non-living organism, the target feature identifier may also indicate the first appearance, the type of non-living object, the acquisition time of the current frame image, the image sequence number of the current frame image, etc. By adding a target feature identifier for the newly added image feature, after the current frame image becomes a historical image, a new frame image after it can iteratively inherit the target feature identifier of the newly added image feature.

[0096] By marking the target feature identifier for the newly added image feature, on the one hand, it is convenient to compare the newly added image feature with the historical feature identifier of the previous frame image, thereby improving the accuracy of the comparison result. In addition, the target feature identifier marked with the newly added image feature also facilitates the accuracy of subsequent iteration inheritance of a new frame image.

[0097] Furthermore, a specific implementation of identifying a newly added image feature as a living body or a non-living body may include: selecting target data corresponding to the newly added image feature from the data collected by the infrared camera installed in the vehicle; and determining whether the newly added image feature is a living body or a non-living body based on the target data. Since living bodies have temperature, the data captured by the infrared camera will contain data indicating heat.

[0098] It is worth noting that the target feature identification or historical feature identification can only distinguish between living organisms and non-living organisms, without marking the specific object of the non-living organism. Preferably, the target feature identification or historical feature identification indicates what the non-living organism is, such as a mobile phone, notebook, electronic product, USB flash drive, etc.

[0099] The technical solution provided for the above embodiments can analyze whether there is any left behind item in different scenarios (for example, the first scenario: a scenario in which a biological object disappears and one or more non-biological objects appear at the same time; the second scenario: a biological object and a non-biological object appear at the same time, but after the biological object disappears, the non-biological object does not disappear at the same time; the third scenario: a non-biological object appears after the biological object, but after the biological object disappears, the non-biological object does not disappear; the fourth scenario: only non-biological objects are left in the car; the fifth scenario: only biological objects other than those in the driving seat are left), so as to prevent passengers from leaving items in the car.

[0100] In addition, there may be multiple specific implementations of the above step S303.

[0101] Specifically, the first specific implementation of step S303 may include: when the comparison result indicates that the real-time image contains a newly added image feature indicating a non-living object and the target image feature indicating a living object in the real-time image disappears, determining the lost items in the vehicle, and marking the newly added image feature with a newly added feature identifier indicating a non-living object. Among them, the disappearance of the target image feature indicating a living object in the real-time image involved in the first specific implementation may be the disappearance of all the target image features indicating a living object in the vehicle, or the disappearance of the target image feature indicating a living object in the area where the newly added image feature appears. The newly added image feature indicating a non-living object in the real-time image involved in the first specific implementation is relative to the last image taken when the vehicle is just started or before the door is locked. By binding the newly added image feature indicating a non-living object with the disappearance of the target image feature indicating a living object, the lost items in the first scene, the second scene, the third scene, and the fourth scene can be analyzed to improve the accuracy of analyzing the lost items.

[0102] The first specific implementation of step S303 may further include: locating the first historical image in which the newly added image feature appears, and determining the items left behind in the car when the first historical image contains the target image feature. That is, the historical image used in the first specific implementation is the first historical image in which the newly added image feature appears. When there are multiple newly added image features and the first historical images in which the multiple newly added image features appear are different, the first historical image is determined for each newly added image feature. That is, when the newly added image feature appears, the target image feature already exists in the area where the newly added image feature is located, then the newly added image feature may be an item left behind by the organism corresponding to the target image feature.

[0103] The second specific implementation of step S303 may include: determining the identification change status corresponding to the newly added image feature and the identification change status corresponding to the target image feature; determining the items left behind in the vehicle when the identification change status corresponding to the newly added image feature is consistent with the identification change status corresponding to the target image feature or when the start time indicated by the identification change status corresponding to the newly added image feature is after the start time indicated by the identification change status corresponding to the target image feature. The identification change status corresponding to the newly added image feature is consistent with the identification change status corresponding to the target image feature, which means that the appearance position of the newly added image feature is consistent with the appearance position of the target image feature, the appearance time of the newly added image feature is later than the appearance time of the target image feature, and each image sequence with the newly added image feature also has the target image feature. Through this process, it can be determined that the newly added image feature is generated by the target image feature, that is, the newly added image feature may be the feature of the item left behind by the organism corresponding to the target image feature, and the newly added image feature can be considered as the item left behind by the target image feature. Based on this, a non-biological object that appears due to a biological object can be identified, and the non-biological object may be an item left behind by a biological object.

[0104] The third specific implementation of step S303 may include: when the comparison result indicates that the target image feature indicating the biological body disappears in the real-time image, analyzing whether there is a new image feature indicating the non-biological body that appears at the same time as the target image feature in the real-time image, and if so, determining the item left in the vehicle. This process is mainly aimed at the scenario where the biological body accidentally leaves the items it carries with it, such as a mobile phone, on the vehicle during the process of getting off the vehicle, so as to timely remind the passengers who get off the vehicle of the items left behind, thereby improving the passengers' riding experience.

[0105] The third specific implementation of step S303 may further include: comparing the real-time image with the target image where the target image feature appears for the first time, and determining whether there is a new image feature indicating a non-living object that appears at the same time as the disappeared target image feature in the real-time image. This process is mainly aimed at the scenario where a living object puts an item on the vehicle after getting on the vehicle, and puts it back on the vehicle after getting off the vehicle, so as to timely remind the passengers who get off the vehicle of the items they left behind, thereby improving the passengers' riding experience.

[0106] The fourth specific implementation of step S303 may include: when the comparison result indicates that there is no image feature indicating a living organism in the real-time image, the first frame image after the door corresponding to the driver's seat of the vehicle is locked is compared with the first frame image after the door corresponding to the driver's seat of the vehicle was locked the last time, to determine whether the real-time image has image features of non-living organisms; when there are image features of non-living organisms, determine the items left in the vehicle. By comparing the first frame image after the door is locked with the first frame image after the door corresponding to the driver's seat of the vehicle was locked the last time, that is, comparing the image after the driver gets on the vehicle with the image after the driver gets off the vehicle, it can be found whether there are items left behind by the driver, and items can be prevented from being left in the vehicle after all passengers and the driver get off the vehicle.

[0107] The fourth specific implementation of step S303 may include: when the comparison result indicates that there is no image feature indicating a living organism in the real-time image, the first frame image after the door corresponding to the driver's seat is locked is directly iteratively inherited from the historical feature identifier of the previous frame image to determine whether there is an image feature of a non-living organism; when there is an image feature of a non-living organism, determine the items left in the car. This technical means is to determine whether there is an image feature of a non-living organism by iteratively inheriting the historical feature identifier of the previous frame image, so that it can quickly find out whether there are items left behind by the driver, and can prevent items from being left in the car after all passengers and the driver get off the car.

[0108] Furthermore, the above-mentioned method for preventing the item from being left behind in the vehicle further includes: matching a target reminder mode for the item left behind in the vehicle from among the preset multiple reminder modes according to the comparison result and the trigger conditions corresponding to the multiple reminder modes. Accordingly, the specific implementation of the above-mentioned step S204 may include: sending a reminder message indicating the item left behind to the information receiving end through the target reminder mode.

[0109] Among them, the reminder mode can be vehicle horn, HUD on the window display of left items, HUD on the front windshield display of left items, passenger's mobile terminal prompt of left items, driver's mobile terminal prompt of left items, etc.

[0110] Among them, the trigger condition can be that the door corresponding to the location of the left-behind item is open, there is no living organism in the car and the door is locked, etc. In addition, the user can also set other trigger conditions according to needs. There is no direct correlation between the above analysis of whether the item is left behind and the trigger condition (that is, the technical means of analyzing whether there is a left-behind item does not determine the trigger condition, and the trigger condition will not reversely limit the technical means of analyzing whether there is a left-behind item). Although there is no direct correlation between the analysis of whether the item is left behind and the trigger condition, by combining the trigger condition with the above analysis technical means, a reminder can be issued after the trigger condition is met, avoiding invalid reminders all the time, and further improving the passenger's riding experience.

[0111] It is worth noting that the above-mentioned image features or feature identifiers indicating image features are based on the first image taken when the door corresponding to the driver's seat is locked and the driver appears in the driving seat, and the image features and feature identifiers of subsequent frames are obtained by iterative inheritance. That is to say, all image features on the first image taken when the door corresponding to the driver's seat is locked and the driver appears in the driving seat are subtracted from each frame as the background, and are not considered in the subsequent frames.

[0112] The following takes the case where the feature identifiers of the image features contained in each frame of image are marked by iterative inheritance as an example, and describes in detail how the technical means provided by the embodiments of the present invention can help the occupants analyze whether there are any left behind items and whether to remind the occupants in different scenarios.

[0113] For the first scenario: Figure 4 As shown, after the driver gets in the car, he puts the folder 40 he carries with him on the co-pilot seat. The image before the driver sits in the driver's seat after the door on the driver's seat is opened is used as the basic image. The subsequent images need to first deduct the image features in the image before the folder 40 appears. On this basis, the technical solution provided by the embodiment of the present invention is aimed at the processing means of the first scenario, such as Figure 8 As shown, the technical solution for the first scenario may include the following steps:

[0114] Step S801: Obtain the current frame image, and obtain the current frame image captured by the infrared cameras corresponding to the driver's seat and the co-driver's seat.

[0115] Step S802: marking a target feature identifier for the image features of the current frame image, and the target feature identifier iteratively inherits the feature identifier marked by the previous frame image.

[0116] For example, after the door on the driver's seat is opened, the image before the driver sits in the driver's seat is used as the basic image. If the image features corresponding to the driver and the image features corresponding to the folder 40 appear in a frame image a after the basic image, the image features of the driver and the image features of the folder 40 are marked with feature identifiers in the image a, and the image features of the driver and the image features of the folder 40 are marked with the shooting time of the image a and the image sequence number of the image a. In a frame image b after the image a, if the image features of the driver and the image features of the folder 40 still exist, the image features of the driver and the image features of the folder 40 in the image b inherit the feature identifier for the image features of the folder 40 marked by the image a, as well as the shooting time of the image a and the image sequence number of the image a. If the image features for the folder 40 are not recognized in the subsequent image c, the image feature mark for the folder 40 is cancelled, and the feature identifier for the image features of the driver is maintained, and the shooting time of the image a and the image sequence number of the image a are maintained with the feature identifier. If the image feature for folder 40 appears again in image d N frames after image c, the feature identifier for the image feature for folder 40 is re-marked in image d, and the shooting time of image c and the image sequence number of image c are marked for the image feature of folder 40. That is, the image feature of folder 40 is used as a newly added image feature in image d, that is, each frame of image iteratively inherits the mark of the previous frame of image, and will not trace back to the previous image, so as to simplify the image marking and effectively improve the accuracy of analyzing whether an item is left behind.

[0117] Step S803: Analyze whether there is any lost item.

[0118] This step may be to compare the target feature identifier of the image feature marker of the current frame image with the feature identifier of the previous frame image marker; or to compare the target feature identifier of the image feature marker of the current frame image with the base image. Or, the image feature marker of the current frame image may be compared with the image in which the image feature marker first appears.

[0119] Step S804: The triggering condition is met and a reminder is triggered.

[0120] For the first scenario, the trigger condition is that the organism (i.e., the driver) in the driving seat disappears. The trigger reminder can be to trigger the vehicle to honk, trigger the HUD on the window glass corresponding to the driving seat to display a folder logo for the above-mentioned folder 40, or send a prompt message to the driver's terminal device.

[0121] For the second scenario: Figure 5As shown, the USB flash drive 50 fell out of the pocket of the passenger in the rear seat after getting out of the car. For this scenario, the image features of the passenger in the image disappear while the image features of the USB flash drive 50 appear. For this scenario, the image after the door on the driver's seat is opened and before the driver sits in the driver's seat can be used as the basic image, or the first image after the door corresponding to the rear seat is opened can be used as the basic image. Subsequent images need to first deduct the image features on the basic image. On this basis, the technical solution provided by the embodiment of the present invention is aimed at the processing means of the second scenario, such as Fig. 9 As shown, the technical solution for the second scenario may include the following steps:

[0122] Step S901: Acquire the current frame image corresponding to the rear seat, and acquire the current frame image captured by the infrared camera corresponding to the rear seat.

[0123] Step S902: marking a target feature identifier for the image feature of the current frame image corresponding to the rear seat, and the target feature identifier iteratively inherits the feature identifier marked in the previous frame image.

[0124] For example, after the door corresponding to the rear seat is opened, the image before the occupant sits on the seat is used as the basic image. If an image feature corresponding to the occupant appears in a frame image a' after the basic image, the image feature of the occupant is marked with a feature identifier, and the image feature of the occupant is marked with the shooting time of image a' and the image serial number of image a'. In a frame image b' after image a', if the image feature of the occupant still exists, the image feature of the occupant in image b' inherits the feature identifier for the image feature of the occupant marked by image a', as well as the shooting time of image a' and the image serial number of image a'. If the image feature for the occupant disappears in the subsequent N frames of image d' and the image feature of the USB flash drive 50 is added, the feature identifier corresponding to the image feature of the USB flash drive 50 and the shooting time of image d' and the image serial number of image d' are marked.

[0125] Step S903: Analyze whether there is any item left behind in the rear seats.

[0126] This step may be to compare the target feature identifier of the image feature marker of the current frame image with the feature identifier of the previous frame image marker; or to compare the target feature identifier of the image feature marker of the current frame image with the base image. Or, the image feature marker of the current frame image may be compared with the image in which the image feature marker first appears.

[0127] Step S904: The triggering condition is met and a reminder is triggered.

[0128] For the second scenario, the trigger condition is that the organism (i.e., the occupant) on the rear seat disappears and a new characteristic mark appears on the rear seat at the same time. The trigger reminder can be to trigger the HUD on the window glass corresponding to the rear seat to display the USB folder mark for the USB flash drive 50, or send a prompt message to the terminal device of the occupant.

[0129] For the third scenario: Figure 6 As shown, after the occupant in the rear seat has been in the vehicle for a while, a mobile phone 60 appears on the item placement area near the rear seat. For this scenario, it is mainly that some time after the image features of the occupant appear in the image, the occupant puts the mobile phone 60 on the item placement area of ​​the vehicle, but the occupant forgets to take the mobile phone 60 with him when getting off the vehicle. For this scenario, the image taken after the door on the driver's seat is opened and before the driver sits in the driver's seat can be used as the basic image, or the first image after the door corresponding to the rear seat is opened can be used as the basic image. Subsequent images need to first deduct the image features on the basic image. On this basis, the technical solution provided by the embodiment of the present invention is a processing method for this third scenario. The technical solution for the third scenario may include the following processes:

[0130] Get the current frame image corresponding to the rear seat, and get the current frame image captured by the infrared camera corresponding to the rear seat. Mark the image features of the current frame image corresponding to the rear seat with a target feature identifier, which iteratively inherits the feature identifier marked by the previous frame image. Analyze whether there are any leftover items in the rear seat. If the trigger condition is met, trigger a reminder.

[0131] The above process is similar to the above Fig. 9 The steps shown are similar. The difference is that, for example, the image feature of the mobile phone 60 appears in an image b' after the image a'. If the image feature of the occupant still exists, the image feature of the occupant in the image b' inherits the feature identifier for the image feature of the occupant marked in the image a', as well as the shooting time of the image a' and the image serial number of the image a', and the image feature identifier of the image feature of the mobile phone 60 in the image b' is marked, as well as the shooting time of the image b' and the image serial number of the image b'. If the image feature for the occupant disappears in the N frames of image d' thereafter, but the feature identifier for the image feature of the mobile phone 60 still exists, the feature identifier for the image feature of the mobile phone 60 is inherited in the image d'.

[0132] In the process of analyzing whether an item is left behind, the image feature mark of the current frame image is compared with the image in which the image feature mark appears for the first time. Since the image features of the mobile phone 60 and the image features of the occupant appear in the same image, it is considered that the mobile phone 60 was placed in the car by the occupant. For the third scenario, the trigger condition is that the organism (i.e., the occupant) on the rear seat disappears and the items that appear in the middle do not disappear at the same time. The trigger reminder can be to trigger the HUD on the window glass corresponding to the rear seat to display the mobile phone logo for the above-mentioned mobile phone 60, or send a prompt message to the occupant's terminal device.

[0133] For the fourth scenario: Figure 7 As shown, passenger A in the last row gets on the bus first and places the notebook 70 on the seat in front of him. Then passenger B gets on the seat next to the notebook 70, but the notebook 70 is not placed by passenger B on the seat next to him. After riding for a while, passenger B gets off the bus first, and then passenger A gets off after a while. For this scenario, the image taken after the door on the driver's seat is opened and before the driver sits in the driver's seat can be used as the basic image, or the first image taken after the door corresponding to the rear seat is opened can be used as the basic image. The image features on the basic image need to be deducted from the subsequent images first. On this basis, the technical solution provided by the embodiment of the present invention is a processing method for the third scenario. The technical solution for the fourth scenario may include the following processes:

[0134] For each rear seat, obtain the current frame image corresponding to the rear seat, and obtain the current frame image captured by the infrared camera corresponding to the rear seat. Mark the image features of the current frame image corresponding to the rear seat with a target feature identifier, which iteratively inherits the feature identifier marked by the previous frame image. Analyze whether there are any leftover items in the rear seat. If the trigger condition is met, trigger a reminder.

[0135] The above process is similar to the above Fig. 9 The steps shown are similar. The difference is, for example, in the frame image b" after image a", the image features of notebook 70 appear, and the image features of occupant A on the last row of seats appear synchronously. After a period of time, the image features of occupant B appear in the captured image c". By comparing the current frame image with the previous frame image or the current frame image with image b" where notebook 70 appears, the image features of notebook 70 can be determined. The fact that it appears before occupant B indicates that notebook 70 does not belong to occupant B. Then, when occupant B disappears, but since it is analyzed that notebook 70 does not belong to occupant B, the triggering reminder condition is not met and no reminder is issued. After occupant A disappears, since the image features of notebook 70 appear at the same time as the image features of occupant A, it is analyzed that notebook 70 is an item left behind by occupant A, and a reminder is triggered.

[0136] Furthermore, the technical solution provided by the embodiment of the present invention can also issue a prompt message when only a biological body belonging to a child or a pet is left in the rear seat and the driver's characteristics in the driving seat disappear.

[0137] In addition, for the technical solutions provided in the above-mentioned embodiments, in addition to comparing images from the same infrared camera, images of each infrared camera can also be compared. In addition, the images synchronously captured by each infrared camera can also be spliced ​​into an image that can reflect the overall situation inside the car. The above-mentioned analysis of whether there is any item left in the car is performed based on the image that can reflect the overall situation inside the car, so as to improve the accuracy of the analysis results.

[0138] Further, Fig.10 FIG. 2 is a schematic diagram showing a partial structure of a device for preventing items from being left behind in a vehicle provided by an embodiment of the present invention. Fig.10 As shown, the in-vehicle article anti-lost reminder device 1000 may include: a feature extraction module 1001, a processing module 1002 and an interaction module 1003, wherein:

[0139] The feature extraction module 1001 is used to extract the first image feature of the current frame image captured by the in-vehicle camera;

[0140] The processing module 1002 is used to compare the first image feature of the current frame image with the historical feature identifier marked by the historical image collected by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image; and analyze whether an item is left in the vehicle according to the comparison result;

[0141] The interaction module 1003 is used to send a prompt message indicating the left-behind item to the information receiving end to remind the occupants when the item is left behind in the vehicle.

[0142] Furthermore, the feature extraction module 1001 is further used to mark a target feature identifier for the first image feature of the current frame image;

[0143] The processing module 1002 is further used to compare the target feature identifier of the current frame image mark with the historical feature identifier of the historical image mark.

[0144] Furthermore, the processing module 1002 is further used to compare the first image feature of the current frame image with the historical feature identifier marked by the previous frame image captured by the in-vehicle camera and the identifier change status corresponding to the historical feature identifier, wherein the historical feature identifier marked by the previous frame image iteratively inherits the historical feature identifiers of the previous multiple frames of images and the identifier change status corresponding to the historical feature identifier; based on the comparison result, the target feature identifier and the identifier change status corresponding to the target feature identifier are marked for the first image feature of the current frame image.

[0145] Furthermore, the feature extraction module 1001 is further used to identify, for each newly added image feature, whether the newly added image feature is a living object or a non-living object, and mark the newly added image feature with a target feature identifier indicating a living object or a non-living object when the comparison result indicates that the real-time image contains one or more newly added image features.

[0146] Furthermore, the feature extraction module 1001 is further used to filter target data corresponding to the newly added image features from the data collected by the infrared camera installed in the vehicle; and determine whether the newly added image features are biological or non-biological objects based on the target data.

[0147] Furthermore, the processing module 1002 is further used to determine the lost items in the vehicle and mark the new image features with new feature identifiers indicating non-living objects when the comparison result indicates that the real-time image contains new image features indicating non-living objects and the target image features indicating living objects in the real-time image disappear.

[0148] Furthermore, the processing module 1002 is further used to locate the first historical image in which the newly added image feature appears, and determine the item left behind in the vehicle when the first historical image contains the target image feature.

[0149] Furthermore, the processing module 1002 is further used to determine the identification change status corresponding to the newly added image feature and the identification change status corresponding to the target image feature; when the identification change status corresponding to the newly added image feature is consistent with the identification change status corresponding to the target image feature or when the start time indicated by the identification change status corresponding to the newly added image feature is after the start time indicated by the identification change status corresponding to the target image feature, the left-behind item in the vehicle is determined.

[0150] Furthermore, the processing module 1002 is further used to analyze whether there is a new image feature indicating a non-living object that appears simultaneously with the target image feature in the real-time image when the comparison result indicates that the target image feature indicating the living object in the real-time image disappears, and if so, determine that the item is left behind in the vehicle.

[0151] Furthermore, the processing module 1002 is further used to compare the real-time image with the target image in which the target image feature appears for the first time, to determine whether there is a new image feature indicating a non-living object that appears simultaneously with the disappeared target image feature in the real-time image.

[0152] Furthermore, the processing module 1002 is further used to compare the first frame image after the door corresponding to the driver's seat of the vehicle is locked with the first frame image after the door corresponding to the driver's seat of the vehicle was locked the last time, when the comparison result indicates that there are no image features indicating a living organism in the real-time image, to determine whether there are image features of non-living organisms in the real-time image; if there are image features of non-living organisms, determine whether the items have been left behind in the vehicle.

[0153] Furthermore, the processing module 1002 is further used to iteratively inherit the historical feature identification of the previous frame image corresponding to the first frame image after the door corresponding to the driver's seat is locked, to determine whether there are image features of non-living objects when the comparison result indicates that there are no image features indicating any living organisms in the real-time image; if there are image features of non-living objects, determine the items left in the car.

[0154] Furthermore, the interaction module 1003 is further used to match a target reminder mode for the left-behind items in the vehicle from the preset multiple reminder modes according to the comparison results and the trigger conditions corresponding to the multiple reminder modes; and send a reminder message indicating the left-behind items to the information receiving end through the target reminder mode.

[0155] Furthermore, an embodiment of the present invention also provides a vehicle item anti-lost reminder system. Fig.11 As shown, the in-vehicle article anti-lost reminder system 1100 may include: an in-vehicle camera 1101 and the in-vehicle article anti-lost reminder device 1000 provided in the above embodiment, wherein:

[0156] The in-vehicle camera 1101 sends each frame of collected images to the in-vehicle item anti-lost reminder device 1000 in real time.

[0157] There are multiple in-vehicle cameras 1101 , which are distributed on the roof of the vehicle and correspond to each seat area respectively.

[0158] Furthermore, the above-mentioned in-vehicle item anti-lost reminder system 1100 may also include: infrared cameras 1102 corresponding to each in-vehicle camera, wherein the image area captured by each infrared camera 1102 is consistent with the image area captured by its corresponding in-vehicle camera.

[0159] Furthermore, an embodiment of the present invention also provides an electronic device. The electronic device may include:

[0160] one or more processors;

[0161] a storage device for storing one or more programs,

[0162] When the one or more programs are executed by the one or more processors, the one or more processors implement the method for preventing items from being left behind in a vehicle as provided in the above embodiment.

[0163] Furthermore, an embodiment of the present invention also provides a computer-readable medium on which a computer program for implementing a method for preventing items from being left behind in a vehicle is stored.

[0164] When the computer program is executed by the vehicle-mounted processor, the method for preventing items from being left behind in the vehicle provided in the first aspect embodiment described above is implemented.

[0165] Furthermore, an embodiment of the present invention also provides a vehicle, which executes the above-mentioned method for reminding people not to lose items in the vehicle.

[0166] Reference below Fig.12 , which shows a schematic diagram of the structure of a computer system 1200 suitable for implementing the method for preventing items from being left behind in a vehicle according to an embodiment of the present invention. Fig.12 The computer system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0167] like Fig.12 As shown, the computer system 1200 includes a central processing unit (CPU) 1201, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage part 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for the operation of the system 1200 are also stored. The CPU 1201, the ROM 1202, and the RAM 1203 are connected to each other via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0168] The following components are connected to the I / O interface 1205: an input section 1206 including an output section 1207 including a cathode ray tube (CRT), a liquid crystal display (LCD), a speaker, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 1209 including a network interface card such as a LAN card, a modem, etc. The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the I / O interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1210 as needed, so that a computer program read therefrom is installed into the storage section 1208 as needed.

[0169] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1209, and / or installed from the removable medium 1211. When the computer program is executed by the central processing unit (CPU) 1201, the above-mentioned functions defined in the system of the present invention are executed.

[0170] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0171] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0172] The modules involved in the embodiments of the present invention may be implemented by software or hardware. The modules described may also be set in a processor. For example, they may be described as: a processor including the above-mentioned feature extraction module, processing module and interaction module. The names of these modules or units do not, in some cases, constitute limitations on the modules or units themselves. For example, the feature extraction module may also be described as "a module or unit for extracting the first image feature of the current frame image captured by the in-vehicle camera".

[0173] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: extracting the first image feature of the current frame image captured by the in-vehicle camera in real time; comparing the first image feature of the current frame image with the historical feature identifier of the historical image captured by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image; analyzing whether there is any item left in the car based on the comparison result; in the case of any item left in the car, sending a prompt message indicating the left item to the information receiving end to remind the occupant.

[0174] According to the technical solution of an embodiment of the present invention, by extracting the first image feature of the current frame image captured in real time by the in-vehicle camera, and comparing the first image feature of the current frame image with the historical feature identifier marked by the historical image captured by the in-vehicle camera, since the historical feature identifier indicates the second image feature included in the historical image, the difference between the first image feature and the historical feature identifier can be quickly obtained by comparing the first image feature with the historical feature identifier, so as to quickly analyze whether an item is left behind from the comparison result. In the case of left behind items in the car, a prompt message indicating the left behind items is sent to the information receiving end to remind the occupants. Therefore, the technical solution provided by the present invention can remind the occupants or the driver of left behind items in the car and improve the riding experience.

[0175] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for preventing items from being left behind in a vehicle, characterized in that: include: Extracting the first image feature of the current frame image collected in real time by the in-vehicle camera; Comparing the first image feature of the current frame image with the historical feature identifier marked in the historical image collected by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image; According to the comparison results, analyze whether there are any items left in the car; In the case of items left behind in the vehicle, a prompt message indicating the left-behind item is sent to the information receiving end to remind the occupants.

2. The method for preventing items from being left behind in a vehicle according to claim 1, characterized in that: The vehicle article anti-lost reminder method further includes: marking a target feature identifier for the first image feature of the current frame image; The comparing the first image feature of the current frame image with the historical feature identifier of the historical image captured by the in-vehicle camera includes: The target feature identifier of the current frame image mark is compared with the historical feature identifier of the historical image mark.

3. The method for preventing items from being left behind in a vehicle according to claim 1, characterized in that: The comparing the first image feature of the current frame image with the historical feature identifier of the historical image captured by the in-vehicle camera includes: Compare the first image feature of the current frame image with the historical feature identifier marked by the previous frame image captured by the in-vehicle camera and the identifier change status corresponding to the historical feature identifier, wherein the historical feature identifier marked by the previous frame image iteratively inherits the historical feature identifiers of the previous multiple frames of images and the identifier change status corresponding to the historical feature identifier; According to the comparison result, a target feature identifier and an identifier change status corresponding to the target feature identifier are marked for the first image feature of the current frame image.

4. The method for preventing items from being left behind in a vehicle according to claim 2, characterized in that: The step of marking a target feature identifier for a first image feature of the current frame image includes: When the comparison result indicates that the real-time image contains one or more newly added image features, For each of the newly added image features, the newly added image feature is identified as a living object or a non-living object, and a target feature identifier indicating a living object or a non-living object is marked on the newly added image feature.

5. The method for preventing items from being left behind in a vehicle according to claim 4, characterized in that: The identifying the newly added image feature as a living object or a non-living object includes: Filtering target data corresponding to the newly added image features from data collected by an infrared camera installed in the vehicle; According to the target data, it is determined that the newly added image feature is a living object or a non-living object.

6. The method for preventing items from being left behind in a vehicle according to any one of claims 1 to 3, characterized in that: The analyzing whether there is any item left in the vehicle includes: When the comparison result indicates that the real-time image includes a newly added image feature indicating a non-living object and the target image feature indicating a living object in the real-time image disappears, determining that the object is left behind in the vehicle and marking the newly added image feature with a newly added feature identifier indicating a non-living object; Preferably, the analyzing whether there is any item left in the vehicle further includes: Locating the first historical image in which the newly added image feature appears, and determining the item left behind in the vehicle when the first historical image contains the target image feature; or, The analyzing whether there is any item left in the vehicle also includes: Determine a change status of a mark corresponding to the newly added image feature and a change status of a mark corresponding to the target image feature; When the identification change status corresponding to the newly added image feature is consistent with the identification change status corresponding to the target image feature or when the start time indicated by the identification change status corresponding to the newly added image feature is after the start time indicated by the identification change status corresponding to the target image feature, the left-behind item in the vehicle is determined.

7. The method for preventing items from being left behind in a vehicle according to any one of claims 1 to 3, characterized in that: The analyzing whether there is any item left in the vehicle includes: When the comparison result indicates that the target image feature indicating the biological object in the real-time image disappears, analyzing whether there is a new image feature indicating the non-biological object that appears simultaneously with the target image feature in the real-time image, and if so, determining that the object is left behind in the vehicle; Preferably, the analyzing whether there is a newly added image feature indicating a non-living object that appears simultaneously with the target image feature in the real-time image comprises: The real-time image is compared with the target image in which the target image feature first appears, to determine whether the real-time image has a newly added image feature indicating a non-living object that appears simultaneously with the disappeared target image feature.

8. The method for preventing items from being left behind in a vehicle according to any one of claims 1 to 3, characterized in that: The analyzing whether there is any item left in the vehicle includes: In the case where the comparison result indicates that there is no image feature indicative of a living organism in the real-time image, Compare the first frame image after the door corresponding to the driver's seat of the vehicle is locked with the first frame image after the door corresponding to the driver's seat of the vehicle was locked the last time, to determine whether the real-time image has image features of non-biological objects; or directly iteratively inherit the historical feature identifier of the previous frame image corresponding to the first frame image after the door corresponding to the driver's seat is locked, to determine whether the image features of non-biological objects exist; Identify items left in a car in the presence of non-biological image features; and / or, The method for preventing objects from being left behind in a vehicle further comprises: matching a target reminder mode for the object left behind in the vehicle from among the preset reminder modes according to the comparison result and the trigger conditions corresponding to the multiple reminder modes; The sending of the prompt information indicating the lost item to the information receiving end includes: sending the prompt information indicating the lost item to the information receiving end through the target reminder mode.

9. A device for preventing items from being left behind in a vehicle, characterized in that: include: Feature extraction module, processing module and interaction module, among which, The feature extraction module is used to extract the first image feature of the current frame image captured by the in-vehicle camera; The processing module is used to compare the first image feature of the current frame image with the historical feature identifier marked by the historical image collected by the in-vehicle camera, wherein the historical feature identifier indicates the second image feature included in the historical image; and analyze whether an item is left in the vehicle based on the comparison result; The interaction module is used to send a prompt message indicating the left-behind item to the information receiving end to remind the occupants when the item is left behind in the vehicle.

10. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the in-vehicle item anti-lost reminder method as described in any one of claims 1 to 8 above.

Images