CN110378216B - Target detection method, target detection device, image pickup apparatus, and storage medium - Google Patents
Target detection method, target detection device, image pickup apparatus, and storage medium Download PDFInfo
- Publication number
- CN110378216B CN110378216B CN201910509107.1A CN201910509107A CN110378216B CN 110378216 B CN110378216 B CN 110378216B CN 201910509107 A CN201910509107 A CN 201910509107A CN 110378216 B CN110378216 B CN 110378216B
- Authority
- CN
- China
- Prior art keywords
- target
- detection
- detection block
- target object
- sensor array
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The application relates to a target detection method, a target detection device, an image pickup apparatus and a storage medium. The method comprises the following steps: the method comprises the steps of carrying out target detection through a preset target detection algorithm in a scene picture shot by a camera on the camera equipment, tracking a target object when the target object is detected, acquiring position information of the target object in the scene picture, determining a target detection block where the position information of the target object is located in a detection block array, detecting the target detection block through a sensor array, and determining the target object detected by the target detection algorithm as an effective target if a biological signal in the target detection block is detected. By adopting the method, the probability of missing detection and false detection of the target detection algorithm can be effectively reduced, and the accuracy of target detection is improved.
Description
Technical Field
The present application relates to the field of sensor and image intelligent analysis technologies, and in particular, to a target detection method and apparatus, an image capturing device, and a storage medium.
Background
The video monitoring is a field in which the implementation of the deep learning algorithm is fast, and is also a field in which the application value of the deep learning algorithm is high. In recent years, a target detection algorithm based on deep learning is a research hotspot in the aspect of deep learning algorithms in the field of video monitoring.
However, due to the shape diversity of the target objects, for example, different shapes of different human targets, different shapes of the same human target in different environments, different shapes of different body behaviors, etc., it is difficult for the training set of the deep learning algorithm to acquire complete positive and negative samples, so that a target detection algorithm based on deep learning may have a lot of false detections and missed detections when performing target detection. At present, false detection and missed detection can be gradually reduced only by updating the training set and retraining the detection network based on the updated training set, which is a relatively passive and long-term improvement scheme.
Disclosure of Invention
In view of the above, it is necessary to provide an object detection method, an apparatus, an image pickup device, and a storage medium capable of effectively improving object detection accuracy in view of the above technical problems.
A target detection method comprises the steps that a sensor array is arranged on a camera device, a single sensor in the sensor array is used for detecting biological signals in a single detection block, and a detection block array is formed by a preset number of detection blocks; the method comprises the following steps:
performing target detection through a preset target detection algorithm in a scene picture shot by a camera on the camera equipment, wherein the scene picture is divided into a plurality of detection blocks;
when a target object is detected, tracking the target object, and acquiring the position information of the target object in the scene picture;
in the detection block array, determining a target detection block where the position information of the target object is located, and detecting the target detection block through the sensor array;
determining the target object as a valid target if the sensor array detects the bio-signal in the target detection zone.
In one embodiment, the step of determining a target detection block in the detection block array where the position information of the target object is located includes:
acquiring the current focal length of the camera;
in the detection block array, determining an effective detection block array corresponding to the current focal length according to a preset mapping relation between the focal length of the camera and the effective detection block array;
and determining a target detection block where the position information of the target object is located in the effective detection block array corresponding to the current focal length.
In one embodiment, the array size of the sensor array is smaller than the detection block array; the step of determining a target detection block in which the position information of the target object is located in the detection block array includes:
determining a detection priority of each target object;
and sequentially determining a target detection block where each target object is located according to the detection priority, and controlling the sensors in the sensor array to move to the target detection block so as to detect the target detection block.
In one embodiment, a rotating motor is connected to the sensor array; the step of controlling the sensors in the sensor array to move to the target detection block in sequence includes:
and controlling the sensor to move to the target detection block through the rotating motor.
In one embodiment, the step of determining the detection priority of each target object includes:
acquiring the entry time, the movement direction and the movement speed of each target object;
predicting the leaving time of each target object according to the entering time, the moving direction and the moving speed;
and determining the detection priority of each target object according to the predicted departure time of each target object.
In one embodiment, the step of controlling the sensors in the sensor array to sequentially move to the target detection block includes:
when the sensor is detected to be in a moving state, interrupting the detection work of the sensor;
and when the sensor is detected to reach the target detection block and stops moving, the detection work of the sensor is resumed.
In one embodiment, the method further comprises:
determining whether a motion trajectory of the target object exists if the sensor array detects that the bio-signal does not exist in the target detection block;
and if the target object has a motion track, determining that the target object is a valid target.
In one embodiment, the method further comprises:
when the target detection algorithm detects that no target object exists in the scene picture, detecting each detection block through the sensor array;
when the biological signal exists in any one detection block, carrying out target detection on the detection block with the biological signal through the target detection algorithm;
determining that the target object detected by the target detection algorithm in the detection block in which the bio-signal is present is a valid target if the target object is detected by the target detection algorithm.
In one embodiment, the method further comprises:
and if the target detection algorithm does not detect the target object in the detection block with the biological signal, adjusting the camera shooting parameters of the camera, and skipping to the step of carrying out target detection on the detection block with the biological signal through the target detection algorithm.
In one embodiment, the sensor array is integrated into a front cover panel of the camera; or, the sensor array is positioned on a separate sensor array plate, and the sensor array plate is matched with the camera for use.
A target detection device is characterized in that a sensor array is arranged on an image pickup device, a single sensor in the sensor array is used for detecting biological signals in a single detection block, and a detection block array is formed by a preset number of detection blocks; the device comprises:
the image acquisition module is used for carrying out target detection through a preset target detection algorithm in a scene image shot by a camera on the camera equipment, and the scene image is divided into a plurality of detection blocks;
the target position acquisition module is used for tracking a target object and acquiring the position information of the target object in the scene picture when the target object is detected;
the target signal detection module is used for determining a target detection block where the position information of the target object is located in the detection block array and detecting the target detection block through the sensor array; and
a valid target determination module for determining the target object as a valid target if the sensor array detects the bio-signal in the target detection zone.
An image pickup apparatus, comprising:
the camera is used for collecting scene pictures;
the sensor array is used for detecting biological signals in a preset detection block array, and the single sensor is used for detecting the biological signals in the single detection block; and
and a processor storing a computer program, the processor executing the computer program to implement the steps of the object detection method.
In one embodiment, a rotary motor is coupled to the sensor array, and movement of the outer sensor array is controlled by the rotary motor.
In one embodiment, the external sensor array is integrated into a front cover panel of the camera; or, the sensor array is positioned on a separate sensor array plate, and the sensor array plate is matched with the camera for use.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the object detection method as described above.
In the target detection method, the device, the computer device and the storage medium, the camera device is provided with the sensor array, the single sensor is used for detecting biological signals of a single detection block, the detection block array is formed by a preset number of detection blocks, target detection is performed through a target detection algorithm, when the target detection algorithm detects a target object, the target detection block where the target object is located is determined, the biological signals in the target detection block are detected through the sensor array, and if the biological signals are detected, the target object detected by the target detection algorithm is determined to be an effective target, so that the target detection is performed by combining the target detection algorithm and the sensor, the probability of missed detection and false detection of the target detection algorithm is effectively reduced, and the accuracy of the target detection is improved.
Drawings
FIG. 1 is a diagram of an exemplary implementation of a target detection method;
FIG. 2 is a schematic flow chart diagram of a method for object detection in one embodiment;
FIG. 3 is an exemplary diagram of a scene in a target detection method in one embodiment;
FIG. 4 is a diagram illustrating an exemplary array of probe blocks in a target detection method according to one embodiment;
fig. 5 is a diagram showing an example of an image pickup apparatus in the target detection method in one embodiment;
FIG. 6 is an exemplary diagram illustrating calibration of motor position in the target detection method in one embodiment;
FIG. 7 is a schematic flow chart diagram of a target detection method in another embodiment;
FIG. 8 is a schematic flow chart diagram of a target detection method in yet another embodiment; and
fig. 9 is a block diagram showing the structure of an object detection device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The target detection method provided by the application can be applied to the application environment shown in fig. 1. The image pickup apparatus 102 communicates with the server 104 via a network. The camera device 102 is used for performing target detection on a scene picture captured by the camera, and the server 104 is used for storing an image of a detected target object, motion trail information and the like. The camera device 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.
In one embodiment, as shown in fig. 2, there is provided an object detection method, which is described by taking an example of application of the method to the image pickup apparatus in fig. 1, and includes the steps of:
The camera device is provided with a sensor array, the sensor array is composed of a plurality of sensors, and the sensors can output voltage change signals when detecting biological signals of a target object. In designing the sensor array, sensor arrays of different sizes, for example 3 × 3, 6 × 6 or 8 × 8 sizes, may be designed according to the detection accuracy requirements of the user. The sensor arrays of size 3 × 3, 6 × 6, 8 × 8 consist of 3 × 3, 6 × 6, 8 × 8 sensors, respectively.
The detection area array is formed by presetting a preset number of detection areas, and each detection area is a detection area of a single sensor. The size of the detection block array is determined by the size of the monitoring area of the camera and the size of the detection area of a single sensor, and can be the same as or larger than the size of the sensor array. For example, the size of the detection block array is M × N, if the value of M × N is small, a sensor array having the same size as the detection block array may be designed, so that the sensors in the sensor array correspond to the detection blocks in the detection block array one by one, and if the value of M × N is large, considering that the implementation of the sensor array having the size of M × N is difficult, a sensor array having an array size smaller than M × N may be designed, and each detection block in the detection block array may be detected by moving the sensor array.
Although the size of the detection block array is fixed, since the field angle of a scene picture shot by the camera is increased or decreased with the change of the focal length of the camera, the scene picture may be only a partial area of an area monitored by the camera, and the scene pictures shot under different focal lengths correspond to the partial detection block arrays with different sizes in the detection block array. For example, the size of the detection block array is 8 × 8, when the field angle of the scene image captured by the camera is the largest, the size of the local detection block array corresponding to the scene image is 8 × 8, and the size of the local detection block array corresponding to the scene image is gradually smaller as the field angle is reduced. When the scene corresponds to the local detection block array of 3 × 3 size, the scene may be considered to be divided into 3 × 3 detection blocks, and when the scene corresponds to the local detection block array of 5 × 5, the scene may be considered to be divided into 5 × 5 detection blocks.
Specifically, a scene picture is shot through a camera on the camera device, a scene video is formed by a plurality of frames of scene pictures, and target detection is performed in each frame of scene picture through a target detection algorithm so as to detect a target object appearing in the scene picture. The target detection algorithm is capable of detecting and tracking a target object in a scene shot by the camera, and is not limited herein, and for example, various target detection algorithms based on deep learning may be used.
And step 204, when the target object is detected, tracking the target object and acquiring the position information of the target object in the scene picture.
Specifically, when a target object is detected in a scene picture by a target detection algorithm, the detected target object is tracked by the target detection algorithm, and the position information of the target object in the scene picture is acquired. The position information of the target object in the scene picture is the position coordinates of the target object in the scene picture, and the target object in the scene picture is usually represented as a corresponding rectangle, so the position coordinates of the target object in the scene picture are the position coordinates of the rectangle.
Specifically, after the position information of the target object detected by the target detection algorithm is obtained, in the detection block array, the detection block in which the position information of the target object is located is determined, and for convenience of distinction, the detection block in which the position information of the target object is located is referred to as a target detection block. And detecting the target detection block by the sensor array.
In step 208, if the sensor array detects the bio-signal in the target detection block, the target object is determined to be a valid target.
Specifically, if the sensor array detects a biological signal in the target detection block, which indicates that the target detection block in which the position information of the target object is located has the target object, it may be determined that the detection result of the target detection algorithm is an effective result, that is, the target object detected by the target detection algorithm is an effective target, so that the detection result of the target detection algorithm is verified by the sensor array, thereby effectively reducing the probability of missed detection and false detection of the target detection algorithm, and improving the accuracy of target detection.
According to the target detection method, the target detection is carried out on the scene picture shot by the camera through the target detection algorithm, when the target detection algorithm detects the target object, the target detection block where the target object is located is determined, the target detection block is detected through the sensor array, and if the biological signals in the target detection block are detected, the target object detected by the target detection algorithm is determined to be an effective target, so that the detection result of the target detection algorithm is verified through the sensor array, the probability of missing detection and false detection of the target detection algorithm is effectively reduced, and the accuracy of the target detection is improved.
In one embodiment, taking fig. 3 as an example, the target object in the scene picture is usually represented as a corresponding rectangle, the position coordinates of the target object in the scene picture, that is, the position coordinates of the rectangle, the target detection algorithm may detect and track multiple target objects at the same time, and each target object may overlap with multiple detection blocks at the same time. At the moment, each target object is numbered to distinguish the target objects, all detection blocks where each target object is located are detected through the sensor array, when all the detection blocks where any one target object is located detect biological signals, the target object is determined to be an effective target, and then each target object can be confirmed, so that the target detection accuracy is effectively improved.
In one embodiment, since a change in a focal length of the camera may cause a change in an angle of view of a scene image captured by the camera, the scene image captured at different focal lengths corresponds to local detection block arrays of different sizes in the detection block array, and when determining a target detection block in which position information of a target object is located, it is only necessary to determine the target detection block in the local detection block array corresponding to a current scene image, where the size of the local detection block array is equal to the size of the current scene image. For convenience of description, the local detection block array corresponding to the scene picture is referred to as an active detection block array. The method comprises the steps of presetting a mapping relation between the focal length of a camera and an effective detection block array, obtaining the current focal length of the camera, inquiring the effective detection block array corresponding to the current focal length in the mapping relation, and determining a target detection block where the position information of a target object is located in the effective detection block array, so that the accuracy of determining the target detection block is effectively improved.
As an example, as shown in fig. 4, the detection block array has a size of 5 × 5, and includes detection blocks 0 to 24, respectively, when the focal length of the camera is large, the camera is in a wide-angle state, the effective detection block array may be a detection block array formed by the detection blocks 0 to 24, and when the focal length of the camera is small, the effective detection block array may be a detection block array formed by the detection blocks 6, 7, 8, 11, 12, 13, 16, 17, and 18, and may also be a detection block array formed by the detection blocks 0, 1, 5, and 6. And designing a relation mapping table between the focal length of the camera and the detection block when setting the mapping relation between the focal length of the camera and the effective detection block array, and calibrating before the equipment leaves a factory.
In one embodiment, when the array size of the sensor array is smaller than the detection block array, the detection priority of each target object is determined, and according to the high-low order of the detection priority, the sensors in the sensor array are controlled to sequentially move to the target detection block where each target object is located to perform detection, so that each target object is sequentially detected when the target detection algorithm detects multiple target objects.
In one embodiment, when performing target detection, the target detection algorithm maintains target object information such as a number, an entry time, an exit time, a movement direction, a movement speed, and the like of each detected target object. The number is the ID of the target object, the entry time is the time when the target object enters the scene picture shot by the camera, the exit time is the time when the target object leaves the scene picture shot by the camera, and the motion direction and the motion speed are the motion speed and the motion direction of the target object in the scene picture shot by the camera. When the detection priority of each target object is determined, the leaving time of each target object can be predicted according to the entering time, the moving direction and the moving speed of each target object, and the detection priority of each target object is determined according to the front and back sequence of the leaving time. The method can set a higher detection priority for the target object which is predicted to leave first, set a lower detection priority for the target object which is predicted to leave later, and detect the target detection area where each target object is located in time, so that the target object is prevented from being missed by the sensor in the detection process.
In one embodiment, the sensor array is connected with a rotating motor, and the rotating motor drives the sensor array to rotate, so that the sensor moves to the target detection block. As shown in fig. 5, the sensor array on the image capturing apparatus is connected to a PT Pan/Tilt (Pan/Tilt, which means a Pan/Tilt that rotates up and down/left and right) so as to drive the sensor array through the PT Pan/Tilt.
In one embodiment, since a plurality of target objects may exist in the scene, when the rotating motor rotates, the sensor may detect a biological signal in a moving state to generate a voltage change, which causes noise to a detection result of the sensor. Therefore, when the sensor is detected to be in a moving state, the detection work of the sensor is interrupted, and when the sensor is detected to reach the target detection block and stops moving, the detection work of the sensor is recovered, so that the detection accuracy of the external sensor is effectively improved.
In one embodiment, when the size of the sensor array is smaller than the detection block array, the sensor array is moved to the corresponding detection block by rotating the motor, so that before use or before leaving a factory, the motor position when the sensor array is moved to each detection block needs to be calibrated to improve the accuracy of the movement of the sensor array.
As an example, as shown in fig. 6, when the size of the sensor array is 2 × 2 and the size of the detection block array is 5 × 5, in calibrating the detection blocks 0 and 6, the target object is arranged to move back and forth at the detection blocks 0 and 6 while driving the rotation motor. When the sensors P0 and P3 in the sensor array generate voltage changes and the sensors P1 and P2 do not generate voltage changes, the position of the rotating motor is recorded, and the calibration of the detection block 0 and the detection block 6 is completed. Therefore, the motor position when the sensor array moves to each detection block can be calibrated.
In one embodiment, the sensor array is a pyroelectric infrared sensor array, the sensor is a pyroelectric infrared sensor, the target object is a human body target, the biological signal is infrared rays emitted by a human body, the pyroelectric infrared sensor can output a voltage change signal when detecting the infrared rays emitted by the human body, the human body targets in different areas can be detected through the pyroelectric infrared sensor array, the target detection algorithm is a human body detection algorithm, and therefore the probability of missing detection and false detection of the human body detection algorithm can be effectively reduced through the pyroelectric infrared sensor array, and the accuracy of human body detection is improved.
In one embodiment, as shown in fig. 7, there is provided an object detection method, which is described by taking an example of application of the method to the image pickup apparatus in fig. 1, and includes the steps of:
And step 704, when the target object is detected, tracking the target object and acquiring the position information of the target object in the scene picture.
Specifically, step 702 and step 704 may refer to the detailed description of step 202 and step 204, which is not repeated herein.
In step 708, if the sensor array detects that the target detection block does not have the biological signal, it is determined whether the target object has a motion track.
Specifically, if the sensor array does not detect the biological signal in the target detection block, the target detection algorithm may perform false detection, that is, the target object detected by the target detection algorithm may be an object similar to the target object in the current environment, for example, when the target object is a human body, the target detection algorithm may perform false detection on a tree similar to the human body as the human body, or the detection of the sensor array may generate an error. Therefore, when the sensor array does not detect the biological signals in the target detection block, the target object is tracked through a target detection algorithm, and whether the target object has a motion track or not is determined.
Specifically, when the target object has a motion track, the detection result of the target detection algorithm can be determined to be an effective result, that is, the target object detected by the target detection algorithm is an effective target, so that the detection result of the target detection algorithm is verified by the sensor array, and the detection result of the sensor array is also verified by the target detection algorithm, so that the probability of missed detection and false detection of the target detection algorithm and the sensor array is effectively reduced, and the accuracy of target detection is improved.
In one embodiment, as shown in fig. 8, there is provided an object detection method, which is described by taking an example of application of the method to the image pickup apparatus in fig. 1, and includes the steps of:
Specifically, the implementation process of step 802 may refer to the detailed description of step 202, and is not described again.
And step 804, detecting each detection block through the sensor array when the target detection algorithm detects that no target object exists in the scene picture.
Specifically, when the target detection algorithm detects that no target object exists in the scene picture, the current focal length of the camera is acquired, the effective detection block array corresponding to the current focal length is inquired, and each detection block in the effective detection block array is detected through the sensor array. If the size of the sensor array is smaller than that of the detection block array, the sensor array sequentially detects each detection block by rotating the motor.
And step 806, when the biological signal is detected in any detection block, performing target detection on the detection block with the biological signal through a target detection algorithm.
Specifically, when the biological signal is detected to exist in any detection block, the target detection algorithm is used for carrying out target detection on the detection block in which the biological signal is detected to further verify the detection result of the sensor array.
Specifically, if the target detection algorithm detects a target object in a detection block in which a biological signal is detected, the detected target object is determined to be an effective target, so that each detection block is detected through the sensor array, the target detection algorithm is assisted to detect the target, and the missing detection efficiency and the false detection efficiency of the target detection algorithm are effectively reduced.
In one embodiment, if the target detection algorithm does not detect the target object in the detection block where the presence of the biological signal is detected, the camera parameters of the camera are adjusted, and the target detection algorithm continues to perform target detection on the detection block where the presence of the biological signal is detected, so as to avoid that the target detection algorithm cannot detect the target object due to the ambient light problem. The camera parameters of the camera can be exposure parameters of the camera.
In one embodiment, when the sensor array is the same size as the detection block array, the sensor array may be integrated on the front cover panel of the camera head or on a separate sensor array board that is used in conjunction with the camera head.
It should be understood that although the various steps in the flow charts of fig. 2-8 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-8 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.
In one embodiment, as shown in fig. 9, there is provided an object detection apparatus 900 comprising: a picture acquisition module 902, a target location acquisition module 904, a target signal detection module 906, and a valid target determination module 908, wherein:
the image acquisition module 902 is configured to perform target detection through a preset target detection algorithm in a scene image shot by a camera on the image pickup apparatus, where the scene image is divided into a plurality of detection blocks;
a target position obtaining module 904, configured to track the target object when the target object is detected, and obtain position information of the target object in the scene picture;
a target signal detection module 906, configured to determine a target detection block where the position information of the target object is located in the detection block array, and detect the target detection block through the sensor array; and
a valid target determination module 908 for determining the target object as a valid target if the sensor array detects the bio-signal in the target detection zone.
In one embodiment, the target signal detection module 906 includes:
the focal length acquisition module is used for acquiring the current focal length of the camera;
the effective array determining module is used for determining an effective detection block array corresponding to the current focal length in the detection block array according to the preset mapping relation between the focal length of the camera and the effective detection block array; and
and the target block determining module is used for determining a target detection block where the position information of the target object is located in the effective detection block array corresponding to the current focal length.
In one embodiment, the array size of the sensor array is smaller than the detection block array; the target signal detection module 906 includes:
the priority determining module is used for determining the detection priority of each target object;
and the target block detection module is used for sequentially determining a target detection block where each target object is located according to the high-low sequence of the detection priority, and controlling the sensors in the sensor array to move to the target detection block so as to detect the target detection block.
In one embodiment, a rotary motor is coupled to the sensor array; the target block detection module includes:
and the motor control module is used for controlling the sensor to move to the target detection block by rotating the motor.
In one embodiment, the priority determination module includes:
the target information acquisition module is used for acquiring the entry time, the movement direction and the movement speed of each target object;
the exit time prediction module is used for predicting the exit time of each target object according to the entry time, the movement direction and the movement speed; and
and the detection priority determining module is used for determining the detection priority of each target object according to the predicted departure time of each target object.
In one embodiment, the target block detection module comprises:
the detection work interruption module is used for interrupting the detection work of the sensor when the sensor is detected to be in a moving state; and
and the detection work recovery module is used for recovering the detection work of the sensor when the sensor is detected to reach the target detection block and stop moving.
In one embodiment, the object detection apparatus 900 further comprises:
the motion track determining module is used for determining whether the target object has a motion track if the sensor array detects that the target detection block does not have the biological signals; and
and the first effective target determining module is used for determining the target object as an effective target if the target object has a motion track.
In one embodiment, the object detection apparatus 900 further comprises:
the polling detection module is used for detecting each detection block through the sensor array when the target detection algorithm detects that no target object exists in the scene picture;
the signal block detection module is used for detecting the detection blocks with the biological signals through a target detection algorithm when the biological signals exist in any detection block; and
a second valid target determination module for determining the target object detected by the target detection algorithm in the detection block in which the bio-signal is present as a valid target if the target object is detected by the target detection algorithm.
In one embodiment, the object detection apparatus 900 further comprises:
and the camera shooting parameter adjusting module is used for adjusting the camera shooting parameters of the camera if the target detection algorithm does not detect the target object in the detection block with the biological signal, and skipping to the step of executing the target detection of the detection block with the biological signal through the target detection algorithm.
For specific limitations of the target detection device, reference may be made to the above limitations of the target detection method, which are not described herein again. The modules in the target detection device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, an image pickup apparatus is provided, which includes a camera, a sensor array, and a processor, in which a computer program is stored, and the processor implements the steps of the above method embodiments when executing the computer program. The camera is used for collecting scene pictures, the sensor array is used for detecting biological signals in a preset detection block array, and the single sensor is used for detecting the biological signals in the single detection block.
In one embodiment, a rotary motor is coupled to the sensor array, whereby movement of the sensor array is controlled by the rotary motor.
In one embodiment, the sensor array is integrated into the front cover panel of the camera; or the sensor array is positioned on a separate sensor array plate, and the sensor array plate is matched with the camera for use.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the above-mentioned method embodiments.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. 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 a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (14)
1. The target detection method is characterized in that a sensor array is arranged on a camera device, a single sensor in the sensor array is used for detecting biological signals in a single detection block, and a detection block array is formed by a preset number of detection blocks; the method comprises the following steps:
performing target detection through a preset target detection algorithm in a scene picture shot by a camera on the camera equipment, wherein the scene picture is divided into a plurality of detection blocks;
when a target object is detected, tracking the target object, and acquiring the position information of the target object in the scene picture;
in the detection block array, determining a target detection block where the position information of the target object is located, and detecting the target detection block through the sensor array; the array size of the sensor array is smaller than that of the detection block array; the step of determining a target detection block in which the position information of the target object is located in the detection block array includes: determining a detection priority of each target object; sequentially determining a target detection block where each target object is located according to the high-low sequence of the detection priority, and controlling the sensors in the sensor array to move to the target detection blocks so as to detect the target detection blocks;
determining the target object as a valid target if the sensor array detects the bio-signal in the target detection zone.
2. The method according to claim 1, wherein the step of determining a target detection block in the detection block array where the position information of the target object is located comprises:
acquiring the current focal length of the camera;
in the detection block array, determining an effective detection block array corresponding to the current focal length according to a preset mapping relation between the focal length of the camera and the effective detection block array;
and determining a target detection block where the position information of the target object is located in the effective detection block array corresponding to the current focal length.
3. The method of claim 1, wherein a rotary motor is coupled to the sensor array; the step of controlling the sensors in the sensor array to move to the target detection block in sequence includes:
and controlling the sensor to move to the target detection block through the rotating motor.
4. The method of claim 1, wherein the step of determining a detection priority for each of the target objects comprises:
acquiring the entry time, the movement direction and the movement speed of each target object;
predicting the leaving time of each target object according to the entering time, the moving direction and the moving speed;
and determining the detection priority of each target object according to the predicted departure time of each target object.
5. The method of claim 1, wherein the step of controlling the sensors in the sensor array to sequentially move to the target detection zone comprises:
when the sensor is detected to be in a moving state, interrupting the detection work of the sensor;
and when the sensor is detected to reach the target detection block and stops moving, the detection work of the sensor is resumed.
6. The method of claim 1, further comprising:
determining whether a motion trajectory of the target object exists if the sensor array detects that the bio-signal does not exist in the target detection block;
and if the target object has a motion track, determining that the target object is a valid target.
7. The method of claim 1, further comprising:
when the target detection algorithm detects that no target object exists in the scene picture, detecting each detection block through the sensor array;
when the biological signal exists in any one detection block, carrying out target detection on the detection block with the biological signal through the target detection algorithm;
determining that the target object detected by the target detection algorithm in the detection block in which the bio-signal is present is a valid target if the target object is detected by the target detection algorithm.
8. The method of claim 7, further comprising:
and if the target detection algorithm does not detect the target object in the detection block with the biological signal, adjusting the camera shooting parameters of the camera, and skipping to the step of carrying out target detection on the detection block with the biological signal through the target detection algorithm.
9. The method of claim 1, wherein the sensor array is integrated into a front cover panel of the camera; or, the sensor array is positioned on a separate sensor array plate, and the sensor array plate is matched with the camera for use.
10. The target detection device is characterized in that a sensor array is arranged on an image pickup device, a single sensor in the sensor array is used for detecting biological signals in a single detection block, and a preset number of detection blocks form a detection block array; the device comprises:
the image acquisition module is used for carrying out target detection through a preset target detection algorithm in a scene image shot by a camera on the camera equipment, and the scene image is divided into a plurality of detection blocks;
the target position acquisition module is used for tracking a target object and acquiring the position information of the target object in the scene picture when the target object is detected;
the target signal detection module is used for determining a target detection block where the position information of the target object is located in the detection block array and detecting the target detection block through the sensor array; the array size of the sensor array is smaller than that of the detection block array; the step of determining a target detection block in which the position information of the target object is located in the detection block array includes: determining a detection priority of each target object; sequentially determining a target detection block where each target object is located according to the high-low sequence of the detection priority, and controlling the sensors in the sensor array to move to the target detection blocks so as to detect the target detection blocks; and
a valid target determination module for determining the target object as a valid target if the sensor array detects the bio-signal in the target detection zone.
11. An image pickup apparatus characterized by comprising:
the camera is used for collecting scene pictures;
the sensor array is used for detecting biological signals in a preset detection block array, and the single sensor is used for detecting the biological signals in the single detection block; and
a processor storing a computer program, the processor executing the computer program to perform the steps in the method of any of claims 1-9.
12. The apparatus of claim 11, wherein a rotary motor is coupled to the sensor array, whereby movement of the sensor array is controlled.
13. The apparatus of claim 11, wherein the sensor array is integrated into a front cover panel of the camera; or, the sensor array is positioned on a separate sensor array plate, and the sensor array plate is matched with the camera for use.
14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910509107.1A CN110378216B (en) | 2019-06-13 | 2019-06-13 | Target detection method, target detection device, image pickup apparatus, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910509107.1A CN110378216B (en) | 2019-06-13 | 2019-06-13 | Target detection method, target detection device, image pickup apparatus, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110378216A CN110378216A (en) | 2019-10-25 |
| CN110378216B true CN110378216B (en) | 2022-03-01 |
Family
ID=68250193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910509107.1A Active CN110378216B (en) | 2019-06-13 | 2019-06-13 | Target detection method, target detection device, image pickup apparatus, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110378216B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110677592B (en) * | 2019-10-31 | 2022-06-10 | Oppo广东移动通信有限公司 | Subject focusing method and device, computer equipment and storage medium |
| CN113392678A (en) * | 2020-03-12 | 2021-09-14 | 杭州海康威视数字技术股份有限公司 | Pedestrian detection method, device and storage medium |
| CN113568497B (en) * | 2020-04-28 | 2023-11-14 | 无锡小天鹅电器有限公司 | Device control method, device, electronic device and computer storage medium |
| CN113115216B (en) * | 2021-02-22 | 2022-09-06 | 浙江大华技术股份有限公司 | Indoor positioning method, service management server and computer storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905706A (en) * | 2014-04-21 | 2014-07-02 | 天彩电子（深圳）有限公司 | Internet protocol camera and control method thereof |
| CN107094223A (en) * | 2017-03-17 | 2017-08-25 | 奥克斯空调股份有限公司 | Obtain the method and device of the high-definition image of intrusion target |
| CN107563961A (en) * | 2017-09-01 | 2018-01-09 | 首都师范大学 | A kind of system and method for the moving-target detection based on camera sensor |
-
2019
- 2019-06-13 CN CN201910509107.1A patent/CN110378216B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103905706A (en) * | 2014-04-21 | 2014-07-02 | 天彩电子（深圳）有限公司 | Internet protocol camera and control method thereof |
| CN107094223A (en) * | 2017-03-17 | 2017-08-25 | 奥克斯空调股份有限公司 | Obtain the method and device of the high-definition image of intrusion target |
| CN107563961A (en) * | 2017-09-01 | 2018-01-09 | 首都师范大学 | A kind of system and method for the moving-target detection based on camera sensor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110378216A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110378216B (en) | Target detection method, target detection device, image pickup apparatus, and storage medium | |
| US10762640B2 (en) | Systems and methods for user detection, identification, and localization within a defined space | |
| CN110998594B (en) | Method and system for detecting motion | |
| CN108447091B (en) | Target positioning method and device, electronic equipment and storage medium | |
| US20070236570A1 (en) | Method and apparatus for providing motion control signals between a fixed camera and a ptz camera | |
| US11514585B2 (en) | Object tracking | |
| JP6806188B2 (en) | Information processing system, information processing method and program | |
| US20060197839A1 (en) | Automatic multiscale image acquisition from a steerable camera | |
| US10979645B2 (en) | Video capturing device including cameras and video capturing system including the same | |
| EP2437115B1 (en) | Camera device with rotary base | |
| US7528881B2 (en) | Multiple object processing in wide-angle video camera | |
| EP3319042A1 (en) | Target tracking device and target tracking method | |
| CN114241016A (en) | Cross-camera track association method and device and electronic equipment | |
| JP6302302B2 (en) | Camera control device | |
| EP3432575A1 (en) | Method for performing multi-camera automatic patrol control with aid of statistics data in a surveillance system, and associated apparatus | |
| CN110941975A (en) | Image acquisition method, angle adjustment device and driving system | |
| EP4354853A1 (en) | Thermal-image-monitoring system using plurality of cameras | |
| JP5270319B2 (en) | Image monitoring device | |
| CN114494847A (en) | Robot, interaction method and device of robot and storage medium | |
| TW200540733A (en) | Device and method for defining an area whose image is monitored | |
| KR102579085B1 (en) | Method and appratus for caturing of semiaquatic animals | |
| KR102587129B1 (en) | Method and appratus for caturing of terrestrial animals | |
| CN112528078B (en) | Image archiving method, device and equipment | |
| CN114060300B (en) | Fan control method and device, storage medium, electronic equipment and fan | |
| KR20090098595A (en) | Artificial intelligence apparatus and method for auto identifying and tracing targets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |