A method, device, computer equipment and storage medium for monitoring high-altitude throwing

Abstract

The application discloses a kind of methods for monitoring high-altitude projectile, device, computer equipment and storage medium, to solve the problem of being identified as high-altitude projectile by non-high-altitude projectile error.The method comprises: when the first acquisition end identifies to reach the abnormal object of expected moving speed, obtain the first video frame containing the abnormal object;Zoom in the first video frame, intercept clear object image;Obtain preset image, the clear image of object including less than expected kinetic energy of kinetic energy on the ground;Contrast whether the object image is consistent with the preset image;If not consistent, then in the video frame collected by the second acquisition end adjacent to the first acquisition end, filter out the second video frame containing the abnormal object;Splice the first video frame and second video frame, obtain panoramic image and motion trajectory information of the appearance of the abnormal object;The object image, the panoramic image and the motion trajectory information are sent to client.

Description

Technical Field

[0001] This invention relates to the field of video processing, and more particularly to a method, apparatus, computer equipment, and storage medium for monitoring objects thrown from heights. Background Technology

[0002] With societal progress, high-rise buildings of all types are springing up everywhere, both in cities and towns. However, this increase in high-rise buildings has also led to a rise in objects being thrown from heights. Throwing objects from heights is not only an uncivilized act but also poses a significant threat to society. In recent years, an increasing number of laws and regulations have begun to hold those responsible for throwing objects from heights accountable.

[0003] To further improve the ability to collect evidence of objects being thrown from heights, an increasing number of high-rise buildings are installing surveillance cameras around their perimeters to monitor for such incidents, facilitating timely warnings and subsequent accountability. In existing technology, these cameras typically determine whether an object has been thrown from a height based on its parabolic trajectory. However, this method can misclassify objects like fallen leaves or rainwater as such, leading to false alarms. These false alarms often send incorrect warning messages, wasting unnecessary manpower and storing numerous redundant false alarm messages in the equipment. Summary of the Invention

[0004] This invention provides a method, apparatus, computer equipment, and storage medium for monitoring objects thrown from heights, in order to solve the problem of misidentifying non-height-thrown objects as high-altitude-thrown objects.

[0005] A method for monitoring objects thrown from high-rise buildings includes:

[0006] When the first acquisition terminal detects an abnormal object that has reached the expected moving speed, it acquires a first video frame containing the abnormal object.

[0007] Zoom in on the first video frame and extract a clear image of the object;

[0008] Acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy;

[0009] Compare whether the object image is consistent with the preset image;

[0010] If they are inconsistent, then the second video frame containing the abnormal object is selected from the video frames acquired by the second acquisition end adjacent to the first acquisition end.

[0011] By stitching together the first video frame and the second video frame, a panoramic image and motion trajectory information of the abnormal object are obtained;

[0012] The object image, the panoramic image, and the motion trajectory information are sent to the client.

[0013] In one possible design, the step of magnifying the first video frame and extracting a clear image of the object includes:

[0014] On the first video frame, the location where the abnormal object appears is obtained, and the object coordinates are obtained;

[0015] Using the object's coordinates as the center point, the first video frame is magnified to the desired multiple to obtain the object image.

[0016] In one possible design, before scaling up the first video frame to the desired magnification, the method further includes:

[0017] Obtain the diameter of the abnormal object;

[0018] Obtain the preset percentage corresponding to the diameter of the object;

[0019] Obtain the aspect ratio of the first video frame;

[0020] The desired multiple is calculated by multiplying the preset percentage by the screen ratio.

[0021] In one possible design, obtaining the diameter of the anomalous object includes:

[0022] Obtain the length and width of the abnormal object;

[0023] Compare the length of the object with the width of the object;

[0024] If the length of the object is greater than the width of the object, then the length of the object shall be taken as the diameter of the object;

[0025] If the length of the object is less than or equal to the width of the object, then the width of the object is taken as the diameter of the object.

[0026] In one possible design, scaling up the first video frame to the desired factor includes:

[0027] Obtain gesture information including swipe direction and swipe length;

[0028] Obtain the horizontal length, vertical length, and diagonal length of the first video frame;

[0029] If the sliding direction is horizontal, the expected multiple is calculated by the horizontal length and the sliding length.

[0030] If the sliding direction is longitudinal, the expected multiple is calculated by the longitudinal length and the sliding length.

[0031] If the sliding direction is oblique, the expected multiple is calculated by the oblique length and the sliding length;

[0032] The first video frame is magnified according to the expected magnification.

[0033] In one possible design, stitching together the first and second video frames to obtain a panoramic image and motion trajectory information of the abnormal object includes:

[0034] Predict the location of the first appearance of the abnormal object to obtain the object's position;

[0035] A third video frame containing the location of the object is acquired from a third acquisition terminal that can capture the image of the object's location.

[0036] The third video frame is used as the location image of the abnormal object;

[0037] By stitching together the first video frame and the second video frame, a panoramic image and a motion trajectory image of the abnormal object are obtained;

[0038] The motion trajectory image and the location image are used as the motion trajectory information.

[0039] In one possible design, before filtering out the second video frame containing the anomalous object, the method further includes:

[0040] Determine whether the second acquisition terminal has acquired the abnormal object;

[0041] If the second acquisition terminal has already acquired the abnormal object, it will not send the object image, the panoramic image, and the motion trajectory information to the client again.

[0042] A device for monitoring objects thrown from high-rise buildings, comprising:

[0043] The first acquisition module is used to acquire a first video frame containing the abnormal object when the first acquisition terminal identifies an abnormal object that has reached the expected moving speed.

[0044] The magnification module is used to magnify the first video frame and extract a clear image of the object.

[0045] The second acquisition module is used to acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy;

[0046] The comparison module is used to compare whether the object image is consistent with the preset image;

[0047] The filtering module is used to filter out the second video frame containing the abnormal object from the video frames captured by the second acquisition terminal adjacent to the first acquisition terminal if there is a discrepancy.

[0048] The stitching module is used to stitch together the first video frame and the second video frame to obtain a panoramic image and motion trajectory information of the abnormal object.

[0049] The sending module is used to send the object image, the panoramic image, and the motion trajectory information to the client.

[0050] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the method for monitoring objects thrown from heights as described above.

[0051] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method for monitoring objects thrown from heights as described above.

[0052] The aforementioned method, apparatus, computer equipment, and storage medium for monitoring objects thrown from heights, when the first acquisition terminal identifies an abnormal object reaching a expected moving speed, acquires a first video frame containing the abnormal object. Then, the first video frame is magnified, and a clear image of the object is extracted. Next, a preset image is acquired, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy. The object image is compared to the preset image; if they do not match, a second video frame containing the abnormal object is selected from the adjacent second acquisition terminal. Finally, the first and second video frames are stitched together to obtain a panoramic image and motion trajectory information of the abnormal object. Finally, the object image, panoramic image, and motion trajectory information are sent to the client. This allows the client to promptly detect abnormal objects detected at the acquisition end. Furthermore, because the acquired abnormal objects are magnified and compared with preset images (which include objects with landing kinetic energy less than expected—objects that might be falsely reported as objects thrown from a height), this process is crucial. These include, but are not limited to, objects with autonomous flight capabilities (flying birds, floating balloons, flying drones, etc.) and falling objects with a weight less than expected (raindrops, fallen leaves, etc.). When an abnormal object does not match the preset object, it is considered a high-altitude object thrown from a height. In this case, the magnified object image, the panoramic image of the object, and the abnormal object's motion trajectory information are sent to the client. This eliminates the possibility of misidentifying non-high-altitude objects as such and issuing false warnings, while also ensuring timely issuance of warnings for objects thrown from a height. This allows the server to accurately filter out objects thrown from heights and issue timely warnings. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A schematic diagram of an application environment for a method for monitoring objects thrown from high altitudes according to an embodiment of the present invention;

[0055] Figure 2 A flowchart illustrating a method for monitoring objects thrown from high-rise buildings according to an embodiment of the present invention;

[0056] Figure 3 A schematic diagram of a device for monitoring objects thrown from high altitudes according to an embodiment of the present invention;

[0057] Figure 4 A schematic diagram of a computer device according to one embodiment of the present invention. Detailed Implementation

[0058] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] The method for monitoring objects thrown from heights provided in this invention can be applied to, for example... Figure 1 In this application environment, both the client and the acquisition end communicate with the server via a network. The server acquires the abnormal object through the acquisition end, obtains a clear image of the object, predicts its location, and finally sends the image and location to the client. The client, also known as the user terminal, is the program that provides local services to the client, corresponding to the server. The client can be installed on, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The acquisition end refers to the device capable of acquiring abnormal objects, including but not limited to cameras and camcorders.

[0060] In one embodiment, such as Figure 2 As shown, a method for monitoring objects thrown from high-rise buildings is provided, which can be applied to... Figure 1 Taking the server in the example, the following steps are included:

[0061] S10: When the first acquisition terminal detects an abnormal object that has reached the expected moving speed, it acquires the first video frame containing the abnormal object.

[0062] In this embodiment, the acquisition end includes a first acquisition end, a second acquisition end, and a third acquisition end, wherein the second acquisition end is adjacent to the first acquisition end. A predetermined movement speed is set. When the first acquisition end detects an abnormal object reaching the predetermined movement speed, it acquires the first video frame of the abnormal object. The predetermined movement speed refers to the object's movement speed in different directions, including but not limited to acceleration, angular velocity, and linear velocity. A video frame refers to a video frame captured by the acquisition end at a specific moment.

[0063] S20: Zoom in on the first video frame and extract a clear image of the object.

[0064] After obtaining the first video frame, a clear image of the object is extracted from the magnified first video frame. Methods for magnifying the video frame include, but are not limited to, magnifying the image centered on the midpoint of the video frame, or irregular magnification based on gestures. Methods for extracting the object image include, but are not limited to, extracting an image at a fixed position with fixed dimensions, or extracting an image with the same aspect ratio as the video frame.

[0065] It is worth noting that since the video frames captured by the acquisition end often contain too much redundant information, it is impossible to obtain accurate information about objects thrown from high altitudes if the video frames are directly compared. Therefore, it is necessary to crop the video frames. In addition, since the objects directly cropped from the video frames are often not clear enough, in step S20, the video frames are first enlarged to crop a clear image of the object that can be used in step S40. This also makes it easier for users to intuitively view the objects thrown from high altitudes after the object image is sent to the corresponding client.

[0066] S30: Acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy.

[0067] After obtaining the object image, a preset image is acquired, which includes a clear image of the object whose landing kinetic energy is less than the expected kinetic energy. Kinetic energy refers to the energy generated by an object's motion; landing kinetic energy refers to the energy generated by the object's motion towards the ground. The object's kinetic energy is affected by its mass and velocity, which includes, but is not limited to, gravitational acceleration, the object's initial velocity, and acceleration. Objects whose landing kinetic energy is greater than the expected kinetic energy include, but are not limited to, birds with autonomous flight capabilities, controlled aircraft, fallen leaves, raindrops, feathers, etc. The expected kinetic energy refers to the critical kinetic energy at which an aerial object is classified as a high-altitude projectile.

[0068] Specifically, when setting a target kinetic energy, if the object's landing kinetic energy is greater than or equal to the target kinetic energy, the object is determined to be a high-altitude projectile; if the object's landing kinetic energy is less than the target kinetic energy, the object is determined not to be a high-altitude projectile. Clear images of objects with kinetic energies less than the target kinetic energy are stored in a database as preset images.

[0069] S40: Compare whether the object image is consistent with the preset image.

[0070] After obtaining the preset image, compare whether the object image is consistent with the preset image. Methods for determining whether the object image is consistent with the preset image include, but are not limited to, polling the pixels in the image and calculating image similarity.

[0071] Specifically, the object image is compared with a preset image. If the object image matches the preset image, the object is determined not to be thrown from a height; if the object image does not match the preset image, the object is determined to be thrown from a height.

[0072] S50: If they are inconsistent, then select the second video frame containing the abnormal object from the video frames acquired by the second acquisition end adjacent to the first acquisition end.

[0073] When the object image is determined to be inconsistent with the preset image, that is, when the abnormal object is determined to be a high-altitude object, the video frame captured by the second acquisition end adjacent to the first acquisition end is obtained, and the second video frame containing the abnormal object is selected from it.

[0074] Specifically, the second acquisition end refers to all acquisition ends that are physically adjacent to the first acquisition end. There may be one or more second acquisition ends, which is not limited here. Video frames are acquired from the second acquisition ends, and the second video frames containing abnormal objects are selected from these video frames.

[0075] S60: Stitch the first video frame and the second video frame to obtain a panoramic image and motion trajectory information of the abnormal object.

[0076] After obtaining the second video frame, the first and second video frames are stitched together. If multiple adjacent second acquisition terminals exist, the second video frame is stitched together with the first video frame according to the physical position of each second acquisition terminal relative to the first acquisition terminal, thereby obtaining a panoramic image of the abnormal object's appearance and its motion trajectory information. The panoramic image refers to the complete environmental image of the abnormal object's appearance, including but not limited to images of all surrounding environments during the abnormal object's movement. The motion trajectory information refers to the trajectory of the abnormal object obtained by stitching together the first and second video frames, which includes but is not limited to the abnormal object's coordinates, image of the abnormal object, and image of the abnormal object's position.

[0077] It is worth noting that due to modernization, existing buildings are becoming increasingly taller, making it difficult to capture complete information about objects thrown from heights using a single acquisition device. This information includes, but is not limited to, panoramic and trajectory data of the thrown object. Therefore, in step S60, different video frames acquired by multiple acquisition devices are stitched together. This allows users to visually view the overall environment of the thrown object when the panoramic image is sent to the client, rather than viewing fragmented images.

[0078] S70: Sends object images, panoramic images, and motion trajectory information to the client.

[0079] The object image, panoramic image, and motion trajectory information are sent to the client, which can be a monitoring terminal or a user terminal that needs to be notified in advance; there is no limitation here.

[0080] Sending object images, panoramic images, and motion trajectory information to the monitoring terminal allows monitoring personnel to respond promptly even when receiving warnings of objects being thrown from heights. Sending this information to other user terminals, including mobile devices, ensures that the warning is promptly disseminated to all pre-defined personnel, preventing injuries caused by objects being thrown from heights.

[0081] It should be noted that this method enables the client to promptly acquire abnormal objects detected at the acquisition end. Furthermore, because the acquired abnormal objects are magnified and compared with a preset image—which includes objects that might be falsely flagged as objects thrown from a height—including objects with autonomous flight capabilities (flying birds, floating balloons, flying drones, etc.) and falling objects with a weight less than expected (raindrops, fallen leaves, etc.), when the abnormal object does not match the preset object, it is considered an object thrown from a height. In this case, the magnified image of the object, along with a panoramic image of the object and its trajectory information, is sent to the client. This eliminates the possibility of mistaking non-objects for objects thrown from a height and issuing false warnings, while also ensuring timely issuance of warnings for objects thrown from a height. This allows the server to accurately filter out objects thrown from heights and issue timely warnings.

[0082] In one embodiment, step S20, namely, zooming in on the first video frame and extracting a clear object image, specifically includes the following steps:

[0083] S21: On the first video frame, obtain the location of the abnormal object and get the object's coordinates.

[0084] S22: Using the object's coordinates as the center point, magnify the first video frame to the desired multiple to obtain the object image.

[0085] In this embodiment, zooming in on the first video frame involves obtaining the location of the abnormal object, i.e., the object's coordinates, on the first video frame. Then, using the object's coordinates as the center point, the first video frame is zoomed in to the desired magnification, and a clear image of the object is extracted according to the aspect ratio of the first video frame.

[0086] Specifically, a planar coordinate system is established with the lower left corner of the first video frame as the origin, and the location where the abnormal object appears is marked as the object's coordinates. A predetermined magnification is set, and the first video frame is magnified with the object's coordinates as the center until the magnification reaches the predetermined magnification. The object image is then extracted according to the aspect ratio of the first video frame.

[0087] It should be noted that, since the object image needs to be compared with the preset image later, it is necessary to obtain the location of the abnormal object in the first video frame and mark the coordinates of that location, which is the object coordinates. To prevent the abnormal object from not being in the magnified image after the video frame is enlarged, it is necessary to obtain the location coordinates in advance and use these coordinates as the center point for video frame magnification to ensure that a clear object image can be obtained later. Then, the object image can be used to confirm whether the abnormal object is a high-altitude object thrown from a height, thereby providing an accurate and efficient warning of objects thrown from a height.

[0088] In one embodiment, before step S22, that is, before the first video frame is magnified to the expected factor, the method further includes the following steps:

[0089] S81: Get the diameter of the abnormal object.

[0090] S82: Get the preset percentage corresponding to the object's diameter.

[0091] S83: Get the aspect ratio of the first video frame.

[0092] S84: Multiply the preset percentage by the screen ratio to calculate the expected multiplier.

[0093] In this embodiment, the diameter of the abnormal object is obtained. This diameter refers to the longest diameter of the abnormal object, which can be either the object's length or width; no limitation is made here. The object diameter and the expected magnification percentage are set to obtain a magnification configuration. Once the object diameter is obtained, the expected percentage corresponding to that diameter is read from the magnification configuration. The aspect ratio of the first video frame is obtained, which includes, but is not limited to, the ratio of the frame's length to its width. The preset percentage is multiplied by the aspect ratio to calculate the expected magnification factor required for the image.

[0094] It should be noted that the expected magnification is calculated based on the expected percentage and video frame to ensure that the final magnified image of the object is clear. This prevents errors caused by insufficient clarity of the object image when comparing it with the preset image, so as to accurately screen out objects thrown from high altitudes. Then, the object image is used to confirm whether the abnormal object is thrown from high altitude, thus providing accurate and efficient warnings for objects thrown from high altitudes.

[0095] In one embodiment, step S81, namely obtaining the diameter of the abnormal object, specifically includes the following steps:

[0096] S811: Get the length and width of the abnormal object.

[0097] S812: Compare the length of an object with the width of an object.

[0098] S813: If the length of an object is greater than its width, then the length of the object shall be used as its diameter.

[0099] S814: If the length of an object is less than or equal to its width, then the width of the object shall be used as its diameter.

[0100] In this embodiment, the length and width of the abnormal object are obtained. The object length can be the horizontal length or the longest possible length of the object; there is no limitation here. The object width can be the vertical width or the shortest possible width of the object; there is no limitation here. Comparing the object length and object width, if the object length is greater than the object width, the object length is taken as the object diameter. If the object length is less than or equal to the object width, the object width is taken as the object diameter.

[0101] It should be noted that before magnifying the first video frame to the expected magnification, the object's length and width are compared to select the object's diameter. This allows for a clear and accurate magnified image of the object, enabling accurate screening of objects thrown from heights. The image of the object is then used to confirm whether the abnormal object is indeed thrown from heights, thus providing accurate and efficient early warnings for objects thrown from heights.

[0102] In one embodiment, if the client can capture user gestures, then in step S22, i.e., the first video frame is magnified to the expected multiple, the first video frame can also be magnified through the following steps:

[0103] S221: Obtain gesture information including swipe direction and swipe length.

[0104] S222: Obtain the horizontal length, vertical length, and diagonal length of the first video frame.

[0105] S223: If the sliding direction is horizontal, the expected multiple is calculated by using the horizontal length and the sliding length.

[0106] S224: If the sliding direction is longitudinal, the expected multiple is calculated by using the longitudinal length and the sliding length.

[0107] S225: If the sliding direction is oblique, the expected multiple can be calculated by using the oblique length and the sliding length.

[0108] S226: Magnify the first video frame according to the expected magnification.

[0109] In this embodiment, the user's input gesture information is obtained on the client side and sent to the server. This gesture information includes, but is not limited to, the swipe direction and swipe length. The swipe direction refers to the direction in which the finger swipes on the client side, including but not limited to horizontal, vertical, and diagonal directions, while the swipe length refers to the distance the finger swipes on the client side.

[0110] Obtain the horizontal, vertical, and diagonal lengths of the first video frame. The horizontal length refers to the horizontal dimension of the first video frame, the vertical length refers to the vertical dimension of the first video frame, and the diagonal length refers to the diagonal length of the first video frame. If the sliding direction is horizontal, the expected multiplier is calculated using the horizontal and vertical lengths, i.e., the ratio of the sliding length to the horizontal length. If the sliding direction is vertical, the expected multiplier is calculated using the vertical and vertical lengths, i.e., the ratio of the vertical and vertical lengths, i.e., the ratio of the vertical and vertical lengths, i.e., the ratio of the vertical and vertical lengths, i.e., the ratio of the diagonal ...

[0111] After obtaining the desired magnification, the first video frame is magnified to the desired magnification.

[0112] It should be noted that steps S221-S226 actually provide a way to obtain object images through client gestures, so that the server can better obtain clear object images according to user needs, accurately and efficiently filter out objects thrown from high altitudes, and then confirm whether the abnormal object is thrown from high altitudes through the object image, thereby providing accurate and efficient warnings for objects thrown from high altitudes.

[0113] In one embodiment, step S60, which involves stitching together the first and second video frames to obtain a panoramic image and motion trajectory information of the abnormal object, specifically includes the following steps:

[0114] S61: Predict the location of the first occurrence of the abnormal object and obtain the object's location.

[0115] S62: Obtain a third video frame containing the object's position from a third acquisition terminal that can capture the image of the object's position.

[0116] S63: Use the third video frame as the location image of the anomalous object.

[0117] S64: Stitch the first video frame and the second video frame to obtain a panoramic image and motion trajectory image of the abnormal object.

[0118] S65: Use motion trajectory images and location images as motion trajectory information.

[0119] In this embodiment, the location of the first appearance of the anomalous object is predicted to obtain the object's location. Based on the object's location, a capture area capable of capturing that location is determined. A third capture device capable of capturing this area, i.e., a third capture device capable of capturing image frames of the object's location, is located. From the video frames captured by the third capture device, a third video frame containing the object's location is obtained, and this third video frame is used as the location image of the anomalous object. The first and second video frames are stitched together to obtain a panoramic image and a motion trajectory image of the anomalous object. The motion trajectory image and the location image are used as motion trajectory information. Here, the motion trajectory image refers to the image containing the motion trajectory of the anomalous object, and the location image refers to the image of the location where the anomalous object first appears. Since the location image is the predicted location, it may or may not contain the anomalous object; this is not limited here.

[0120] It should be noted that sending the predicted location image of the object to the client makes the early warning of objects thrown from high-rise buildings more efficient and accurate. This allows monitoring personnel and relevant personnel to more efficiently locate the object thrown from the high-rise building through the location image, facilitating subsequent accountability and improving the overall monitoring process for objects thrown from high-rise buildings.

[0121] In one embodiment, before step S50, i.e. before filtering out the second video frame containing the anomalous object, the method further includes the following steps:

[0122] S91: Determine whether the second acquisition terminal has acquired any abnormal objects;

[0123] S92: If the second acquisition terminal has acquired an abnormal object, it will not send the object image, panoramic image and motion trajectory information to the client again.

[0124] In this embodiment, the second acquisition terminal is one of all acquisition terminals adjacent to the first acquisition terminal in the physical world. Therefore, after identifying the abnormal object, the second acquisition terminal will also send a warning message to the client according to steps S10-S70, that is, send the object image, panoramic image, and motion trajectory information to the client. When the first acquisition terminal identifies the abnormal object, it needs to determine whether the second acquisition terminal has already sent a warning message to the client for the same abnormal object. If the first acquisition terminal sends a warning message again according to steps S10-S70, the client will receive the same warning message repeatedly, causing message redundancy. Therefore, in step S92, if the second acquisition terminal has identified the abnormal object, the first acquisition terminal does not perform the operation of steps S10-S70, that is, it does not repeatedly send the object image, panoramic image, and motion trajectory information to the client.

[0125] It should be noted that adjacent acquisition terminals often identify the same abnormal object. Since the acquisition terminal automatically runs steps S10-S70 after identifying the abnormal object, when there are multiple acquisition terminals in the application environment, there will be multiple repeated alarms for the same abnormal object. This is one of the situations of false alarms for high-altitude object throwing warnings. Since high-rise buildings in traditional solutions often correspond to multiple acquisition terminals, redundant false alarm information often accumulates on the client side. Steps S91-S92 effectively prevent this situation. If other adjacent second acquisition terminals have already identified the abnormal object, the current first acquisition terminal will no longer execute steps S91-S92, effectively and accurately preventing false alarms of high-altitude object throwing, thereby providing accurate and efficient high-altitude object throwing warnings.

[0126] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0127] In one embodiment, a device for monitoring objects thrown from heights is provided, which corresponds one-to-one with the method for monitoring objects thrown from heights in the above embodiments. For example... Figure 3 As shown, the device for monitoring objects thrown from heights includes a first acquisition module 10, a magnification module 20, a second acquisition module 30, a comparison module 40, a filtering module 50, a stitching module 60, and a transmission module 70. Detailed descriptions of each functional module are as follows:

[0128] The first acquisition module 10 is used to acquire a first video frame containing the abnormal object when the first acquisition terminal identifies an abnormal object that has reached the expected moving speed.

[0129] The magnification module 20 is used to magnify the first video frame and extract a clear image of the object.

[0130] The second acquisition module 30 is used to acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy.

[0131] The comparison module 40 is used to compare whether the object image is consistent with the preset image;

[0132] The filtering module 50 is used to filter out the second video frame containing the abnormal object from the video frames acquired by the second acquisition end adjacent to the first acquisition end if there is a discrepancy.

[0133] The stitching module 60 is used to stitch together the first video frame and the second video frame to obtain a panoramic image and motion trajectory information of the abnormal object.

[0134] The sending module 70 is used to send object images, panoramic images, and motion trajectory information to the client.

[0135] Specific limitations regarding the device for monitoring objects thrown from heights can be found in the limitations of the method for monitoring objects thrown from heights mentioned above, and will not be repeated here. Each module in the aforementioned device for monitoring objects thrown from heights can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0136] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores all data generated in the aforementioned method for detecting objects thrown from heights, including first video frames, object images, preset images, and object positions. The network interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for monitoring objects thrown from heights.

[0137] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps:

[0138] When the first acquisition terminal detects an abnormal object that has reached the expected moving speed, it acquires the first video frame containing the abnormal object.

[0139] Zoom in on the first video frame and extract a clear image of the object;

[0140] Acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy;

[0141] Compare whether the object image matches the preset image;

[0142] If they are inconsistent, then the second video frame containing the abnormal object is selected from the video frames acquired by the second acquisition end adjacent to the first acquisition end.

[0143] By stitching together the first and second video frames, a panoramic image and motion trajectory information of the abnormal object are obtained.

[0144] The object image, panoramic image, and motion trajectory information are sent to the client.

[0145] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0146] When the first acquisition terminal detects an abnormal object that has reached the expected moving speed, it acquires the first video frame containing the abnormal object.

[0147] Zoom in on the first video frame and extract a clear image of the object;

[0148] Acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy;

[0149] Compare whether the object image matches the preset image;

[0150] If they are inconsistent, then the second video frame containing the abnormal object is selected from the video frames acquired by the second acquisition end adjacent to the first acquisition end.

[0151] By stitching together the first and second video frames, a panoramic image and motion trajectory information of the abnormal object are obtained.

[0152] The object image, panoramic image, and motion trajectory information are sent to the client.

[0153] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0155] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for monitoring objects thrown from high-rise buildings, characterized in that, include: When the first acquisition terminal detects an abnormal object that has reached the expected moving speed, it acquires a first video frame containing the abnormal object. Zoom in on the first video frame and extract a clear image of the object; Acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy; Compare whether the object image is consistent with the preset image; If they are inconsistent, then the second video frame containing the abnormal object is selected from the video frames acquired by the second acquisition end adjacent to the first acquisition end. By stitching together the first video frame and the second video frame, a panoramic image and motion trajectory information of the abnormal object are obtained; The object image, the panoramic image, and the motion trajectory information are sent to the client. The process of stitching together the first and second video frames to obtain a panoramic image and motion trajectory information of the abnormal object includes: Predict the location of the first appearance of the abnormal object to obtain the object's position; A third video frame containing the location of the object is acquired from a third acquisition terminal that can capture the image of the object's location. The third video frame is used as the location image of the abnormal object; By stitching together the first video frame and the second video frame, a panoramic image and a motion trajectory image of the abnormal object are obtained; The motion trajectory image and the location image are used as the motion trajectory information.

2. The method for monitoring objects thrown from heights as described in claim 1, characterized in that, The step of magnifying the first video frame and extracting a clear image of the object includes: On the first video frame, the location where the abnormal object appears is obtained, and the object coordinates are obtained; Using the object's coordinates as the center point, the first video frame is magnified to the desired multiple to obtain the object image.

3. The method for monitoring objects thrown from heights as described in claim 2, characterized in that, Before scaling up the first video frame to the desired magnification, the method further includes: Obtain the diameter of the abnormal object; Obtain the preset percentage corresponding to the diameter of the object; Obtain the aspect ratio of the first video frame; The desired multiple is calculated by multiplying the preset percentage by the screen ratio.

4. The method for monitoring objects thrown from heights as described in claim 3, characterized in that, The step of obtaining the diameter of the abnormal object includes: Obtain the length and width of the abnormal object; Compare the length of the object with the width of the object; If the length of the object is greater than the width of the object, then the length of the object shall be taken as the diameter of the object; If the length of the object is less than or equal to the width of the object, then the width of the object is taken as the diameter of the object.

5. The method for detecting objects thrown from heights as described in claim 2, characterized in that, The step of magnifying the first video frame to the desired factor includes: Obtain gesture information including swipe direction and swipe length; Obtain the horizontal length, vertical length, and diagonal length of the first video frame; If the sliding direction is horizontal, the expected multiple is calculated by the horizontal length and the sliding length. If the sliding direction is longitudinal, the expected multiple is calculated by the longitudinal length and the sliding length. If the sliding direction is oblique, the expected multiple is calculated by the oblique length and the sliding length; The first video frame is magnified according to the expected magnification.

6. The method for monitoring objects thrown from heights as described in claim 1, characterized in that, Before filtering out the second video frame containing the anomalous object, the method further includes: Determine whether the second acquisition terminal has acquired the abnormal object; If the second acquisition terminal has already acquired the abnormal object, it will not send the object image, the panoramic image, and the motion trajectory information to the client again.

7. A device for monitoring objects thrown from high-rise buildings, characterized in that, include: The first acquisition module is used to acquire a first video frame containing the abnormal object when the first acquisition terminal identifies an abnormal object that has reached the expected moving speed. The magnification module is used to magnify the first video frame and extract a clear image of the object. The second acquisition module is used to acquire a preset image, which includes a clear image of an object whose landing kinetic energy is less than the expected kinetic energy; The comparison module is used to compare whether the object image is consistent with the preset image; The filtering module is used to filter out the second video frame containing the abnormal object from the video frames captured by the second acquisition terminal adjacent to the first acquisition terminal if there is a discrepancy. The stitching module is used to stitch together the first video frame and the second video frame to obtain a panoramic image and motion trajectory information of the abnormal object. The sending module is used to send the object image, the panoramic image, and the motion trajectory information to the client.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for monitoring objects thrown from heights as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method for monitoring objects thrown from heights as described in any one of claims 1 to 6.

Images