Storage cabinet provided with motion sensor

Abstract

The invention discloses a storage cabinet comprising an action sensor. The motion sensor uses a motion detection algorithm to calculate coordinates of an object, or uses an image time difference algorithm to calculate a motion vector of the object, or compares a current image frame with a background image frame so as to judge whether the object is placed in the storage cabinet or taken out of the storage cabinet, and outputs a placing signal or a taking-out signal to a rear-end control system.

Description

Technical Field

[0001] The present invention relates to a storage cabinet equipped with a motion sensor, and more particularly to a storage cabinet that uses a motion sensor to perform motion detection, image temporal difference, and background comparison to determine the moving direction and presence of objects inside the storage cabinet. Background Art

[0002] In a sensor having a pixel matrix, for motion detection, a read circuit reads a first image frame of the pixel matrix at a first time and stores it in a first frame buffer. Then, at a second time, the read circuit reads a second image frame of the pixel matrix and stores it in a second frame buffer.

[0003] When performing motion identification, a processor accesses the first image frame from the first frame buffer and accesses the second image frame from the second frame buffer for calculation.

[0004] That is, the sensor must have at least two frame buffers.

[0005] In view of this, what is needed is a motion sensor that does not need to first store pixel data obtained at different times in frame buffers and then perform pixel operations. Summary of the Invention

[0006] The present invention provides a storage cabinet configured with smart motion detection for determining the position of an object in the storage cabinet based on multiple pixels detecting motion.

[0007] The present invention also provides a storage cabinet equipped with a motion sensor, which is used to determine the moving direction of an object in the storage cabinet according to the motion vector of a pixel region with recognizable brightness change in a temporal difference image frame.

[0008] The present invention also provides a storage cabinet equipped with a motion sensor, which is used to compare a current image frame with a background image frame to determine whether there is an object inside the storage cabinet.

[0009] The present invention provides a storage cabinet including a cabinet body, a cabinet door, and a motion sensor. The cabinet body has an internal space. The cabinet door is used to enclose or open the internal space. The motion sensor is disposed inside the storage cabinet and is used to acquire an image of the internal space when the cabinet door is opened. The motion sensor includes a pixel matrix and a processor. The pixel matrix includes a plurality of pixels arranged in a matrix, and each of the plurality of pixels is used to output temporal difference pixel data. The processor is used to determine a plurality of pixels detecting motion according to the temporal difference pixel data, and determine the object coordinates in the storage cabinet according to the plurality of pixels.

[0010] The present invention also provides a storage cabinet including a cabinet body, a cabinet door, and a motion sensor. The cabinet body has an internal space. The cabinet door is used to enclose or open the internal space. The motion sensor is disposed inside the storage cabinet and is used to obtain an image frame of the internal space when the cabinet door is opened. The motion sensor includes a pixel matrix and a processor. The pixel matrix includes a plurality of pixels arranged in a matrix and is used to output the image frame. The processor is used to judge the motion vector of at least one pixel region where the brightness change exceeds a change threshold according to the time-differenced image frame, and output a direction signal accordingly.

[0011] The present invention also provides a storage cabinet including a cabinet body, a frame buffer, a motion sensor, and a processor. The cabinet body has an internal space. The cabinet door is used to enclose or open the internal space. The frame buffer is used to record a background image frame. The motion sensor is disposed inside the storage cabinet and is used to obtain a current image frame of the internal space. The processor is used to compare the current image frame with the background image frame to judge the objects in the internal space.

[0012] In order to make the above and other objects, features, and advantages of the present invention more obvious, the following will be described in detail in conjunction with the accompanying drawings. In addition, in the description of the present invention, the same components are denoted by the same reference numerals, which are hereby stated in advance. Description of the Drawings

[0013] Figures 1A - 1D is a schematic diagram of the operation of a light sensor according to an embodiment of the present invention;

[0014] Figure 2 is a schematic diagram of the operation of a light sensor according to another embodiment of the present invention;

[0015] Figure 3 is a block schematic diagram of a navigation device according to an embodiment of the present invention;

[0016] Figure 4 is a block schematic diagram of a navigation device according to another embodiment of the present invention;

[0017] Figures 5A to 5C is a schematic diagram of a storage cabinet according to an embodiment of the present invention in different states;

[0018] Figure 6 is a schematic diagram of a motion sensor of a storage cabinet according to an embodiment of the present invention;

[0019] Figure 7 is a schematic diagram of the object coordinates detected by the motion sensor of a storage cabinet according to an embodiment of the present invention at different times;

[0020] Figure 8Schematic diagram of the motion sensor of the storage cabinet according to the embodiment of the present invention detecting a pixel region with recognizable brightness change;

[0021] Figure 9A Circuit diagram of the pixel circuit of the motion sensor capable of outputting time-differential pixel data according to the embodiment of the present invention;

[0022] Figure 9B For Figure 9A Schematic diagram during the operation of the motion sensor;

[0023] Figure 9C Circuit diagram of the differential circuit of the motion sensor capable of outputting time-differential pixel data according to the embodiment of the present invention;

[0024] Figures 10A to 10C Schematic diagrams of the storage cabinet in different states according to other embodiments of the present invention;

[0025] Figures 11A to 11C Schematic diagrams of the storage cabinet in different states according to other embodiments of the present invention;

[0026] Figure 12 Schematic diagram of the motion sensor according to the embodiment of the present invention applied to a trash can;

[0027] Figure 13 Schematic diagram of the motion sensor according to the embodiment of the present invention applied to a storage space; and

[0028] Figure 14 Schematic diagram of the motion sensor according to the embodiment of the present invention applied to a vehicle interior space.

[0029] Description of reference numerals

[0030] 500 Storage cabinet

[0031] 50 Cabinet body

[0032] 51 Cabinet door

[0033] 52 Motion sensor

[0034] 600 Pixel matrix

[0035] 31 Reading circuit

[0036] 37 Timing controller

[0037] 53 Processor

[0038] 1000 Object Detailed implementation manners

[0039] The motion sensor and navigation device according to the embodiments of the present invention respectively determine motion and lift based on pixel data that has sequentially undergone temporal difference and spatial difference. The temporal difference and spatial difference of the present invention are performed during the period of reading pixel data from the pixels, so as to complete the hybrid difference operation at the analog stage.

[0040] Refer to Figure 1A As shown, it is a schematic diagram of an optical sensor (such as a CMOS image sensor, but not limited thereto) according to an embodiment of the present invention. The optical sensor is applicable to, but not limited to, a security system as a motion sensor for sending an activation signal S trig to the host 21 to start a recording program, increase the frame rate of the pixel matrix, turn on a light source, increase the frame size of the image frame output by the pixel matrix, etc., at least one of which, but not limited to this. In some embodiments, when the optical sensor does not detect motion, the host 21, for example, enters a sleep mode or a power-saving mode to reduce power consumption.

[0041] The host 21 is, for example, separately provided from the optical sensor (such as configured in a camera) but coupled to each other to transmit control signals and image data therebetween.

[0042] The optical sensor includes a pixel matrix 100, a reading circuit 11, a comparator 13, a counter 15, and a timing controller 17. Among them, although Figures 1A to 1D the comparator 13 and the counter 15 are shown as independent circuits, the present invention is not limited thereto. In other embodiments, the comparator 13 and the counter 15 may be included in the reading circuit 11 or a processor (not shown) of the optical sensor, where the processor is, for example, a microprocessor (MCU), a digital signal processor (DSP), or an application-specific integrated circuit (ASIC).

[0043] The pixel matrix 100 includes a plurality of pixels ( Figures 1A to 1D illustrated by 8×8 pixels as an example) arranged in a matrix, and each of the plurality of pixels is used to output temporal difference pixel data. The temporal difference pixel data is, for example, the difference in the charge generated by the photodiodes of each of the plurality of pixels at the current time and a reference time (such as setting a new reference time every predetermined time during operation); or the difference in the charge generated by the photodiodes of each of the plurality of pixels at the current time and a previous time one frame period ago.

[0044] A pixel structure that can calculate and output time-difference pixel data is known, and this pixel structure is not the main purpose of the present invention. For example, refer to FIG. 1(a) of "Event-based Vision: A Survey" with the title at https: / / arxiv.org / pdf / 1904.08405, and the entire content of this survey is incorporated herein by reference. The present invention lies in using the time-difference pixel data to achieve motion detection and lift detection.

[0045] The time-difference pixel data depends on the pixel circuit used. In a non-limiting embodiment, when the charge value generated by the photodiode increases beyond a predetermined threshold between two times, a high potential (e.g., represented by "1") is output as the time-difference pixel data; and when the charge value generated by the photodiode decreases below the predetermined threshold between two times, a low potential (e.g., represented by "0") is output as the time-difference pixel data; it can also be the reverse. That is, the time-difference pixel data is the voltage difference caused by the charge generated by the photodiode between two times (e.g., the current time and the reference time or the current time and the previous time).

[0046] The reading circuit 11 sequentially reads the pixel matrix 100 using a reading frame and calculates the spatial difference of the time-difference pixel data between the central pixel of the reading frame and the surrounding pixels of the central pixel.

[0047] For example, Figure 1A in, the reading circuit 11 simultaneously reads the time-difference pixel data of a total of 9 pixels, namely the central pixel P10 of the reading frame A1 and the surrounding pixels P11 to P18 of the central pixel P10, at a first time. The pixel data that the reading circuit 11 can read out is determined according to the control signals of the timing controller 17 (e.g., including row selection signals and reading signals, etc.).

[0048] In a non-limiting embodiment, when the reading circuit 11 reads the 9 time-difference pixel data, it can first calculate the difference between the time-difference pixel data of the central pixel P10 and each of the time-difference pixel data of the surrounding pixels P11 to P18 as a way to calculate the spatial difference. Then, the reading circuit 11 calculates the sum of the 8 absolute values of the 8 obtained differences as the mixed difference of the 9 pixels. The comparator 13 then compares this sum with a predetermined value (e.g., a voltage value, a current value, or a charge / discharge time according to different applications) to determine the brightness change within the range of the reading frame A1. When the mixed difference of the reading frame A1 is greater than or equal to the predetermined value, the count value of the counter 15 is incremented by 1; otherwise, the count value is not incremented.

[0049] Next, the reading frame moves one pixel distance to the right of the pixel matrix 100 and reaches as Figure 1BThe displayed position A2. The reading circuit 11 simultaneously reads the central pixel P20 of the reading frame A2 and the time-difference pixel data of a total of 9 surrounding pixels P21 to P28 of the central pixel P20 at the second time. The sum of the absolute values of the 8 differences calculated by the reading circuit 11 is used as the mixed difference of the 9 pixels in the reading frame A2, which is similar to the calculation of the mixed difference of the reading frame A1 above. The comparator 13 similarly compares the sum of the reading frame A2 calculated by the reading circuit 11 with the predetermined value. The counter 15 similarly increments the count value by 1 when the mixed difference of the reading frame A2 is greater than or equal to the predetermined value; otherwise, the count value is not increased.

[0050] The timing controller 17 sequentially controls the pixels that can be read in the pixel matrix 100 so that the reading circuit 11 sequentially (for example, first to the right and then downwards) reads all the pixels of the pixel matrix 100 until the mixed difference of the reading frame A36 (as Figure 1D shown) is calculated for the comparator 13 to compare with the predetermined value. In this embodiment, the mixed differences related to the reading frames A1 to A36 are the values that have sequentially completed the time and space differences.

[0051] The counter 15 counts / accumulates the number of reading frames in the pixel matrix 100 where the mixed difference is greater than or equal to the predetermined value. When the counted / accumulated number is greater than or equal to a predetermined number (for example, a predetermined proportion of the number of pixels in the pixel matrix 100 as a counting threshold), it indicates that an activity is detected, and the counter 15 then issues an activation signal S trig to the host 21.

[0052] It must be noted that in other embodiments, as long as the counted / accumulated number is greater than or equal to the predetermined number, the scanning of the pixel matrix 100 can be stopped and the counter 15 transmits the activation signal S trig to the host 21. That is, if the predetermined number has been reached in any previous reading frame, it is not necessary for the reading circuit 11 to read until the last reading frame A36.

[0053] Figures 1A to 1D In, only the reading line Lr between the reading circuit 11 and the pixels that can be read is shown to indicate that the time-difference pixel data of these pixels is read by the reading circuit 11 through the reading line Lr. The other pixels that cannot be read are not shown connected to the reading line Lr, thereby simplifying the diagram. Figures 1A to 1D The arrow in indicates the moving direction of the reading frame.

[0054] It must be noted that although Figures 1A to 1DIn [description], the reading frames A1 to A36 are shown as pixel ranges each including 3×3 pixels, but the present invention is not limited thereto. In embodiments where the pixel matrix 100 includes more pixels, the reading frame may include more pixels, such as a pixel range of 5×5 pixels.

[0055] Referring to Figure 2 , in another embodiment, the optical sensor includes six comparators 13. The reading circuit 11 simultaneously reads the pixel data of three rows of the pixel matrix 100 and obtains the hybrid difference of each of the reading frames A1 to A6 using the same method as described above. The comparator 13 then compares the hybrid difference with a predetermined value to determine the brightness change of each of the reading frames A1 to A6. For example, the leftmost comparator 13 compares the hybrid difference related to the reading frame A1 with the predetermined value; the second comparator 13 from the left compares the hybrid difference related to the reading frame A2 with the predetermined value; …; the second comparator 13 from the right compares the hybrid difference related to the reading frame A5 with the predetermined value; and the rightmost comparator 13 compares the hybrid difference related to the reading frame A6 with the predetermined value. The counter 15 then calculates the number of reading frames A1 to A6 in which the hybrid difference is greater than the predetermined value.

[0056] Next, the three rows of pixels read by the reading circuit 11 are each shifted downward by one pixel distance until the last row of pixels of the pixel matrix 100 is read or scanned, and the reading circuit 11, the comparator 13, and the counter 15 repeat the procedure in the previous paragraph. When a predetermined count value is reached (after scanning all or part of the pixels of the pixel matrix 100), the counter 15 issues an activation signal S trig . It can be understood that when the pixel matrix 100 has more columns, more comparators 13 are required.

[0057] Please refer to Figure 3 shown, which is a schematic diagram of a navigation device according to an embodiment of the present invention. The navigation device is, for example, an optical mouse, and its optical sensor (such as a CMOS image sensor, but not limited thereto) includes a first pixel region 300, at least one second pixel region (for example, four second pixel regions 301 to 304 are respectively disposed on four sides of the first pixel region 300 as shown here), a reading circuit 31, a timing controller 37, and a processor 39, wherein the processor 39 includes, for example, a microprocessor (MCU), a digital signal processor (DSP), or an application specific integrated circuit (ASCI). The timing controller 37 is used to generate control signals (such as including row scanning signals and reading signals, etc.) so that the reading circuit 31 can read the pixel data of each pixel.

[0058] It must be noted that although Figure 3The second pixel regions 301 to 304 are shown to have a spaced distance from the first pixel region 300, but the present invention is not limited thereto. In other embodiments, the second pixel regions 301 to 304 are disposed adjacent to the first pixel region 300.

[0059] It must be noted that although Figure 3 the first pixel region 300 and the second pixel regions 301 to 304 are shown to have their pixel data read by the same reading circuit 31, the present invention is not limited thereto. In other embodiments, the first pixel region 300 and the second pixel regions 301 to 304 have their pixel data read by different reading circuits. In one embodiment, the first pixel region 300 and the second pixel regions 301 to 304 are disposed on the same base layer. As Figure 4 shown, the second pixel regions (regions with grid lines shown) and the first pixel region (region shown as blank) are different regions of the same pixel matrix 400.

[0060] The following takes Figure 3 as an example for illustration. Those skilled in the art to which the present invention pertains can still understand Figure 3 the operation mode of Figure 4 after understanding the description of

[0061] The first pixel region 300 is used to output first image data, and the size of the first image data is determined according to the number of pixels included in the pixel matrix of the first pixel region 300. After reading the first pixel region 300, the reading circuit 31 outputs the first image data I move to the processor 39.

[0062] The second pixel regions 301 to 304 are disposed on four sides of the first pixel region 300. Each of the second pixel regions 301 to 304 is respectively used to output second image data. More specifically, after reading the second pixel regions 301 to 304, the reading circuit 31 respectively outputs the second image data I lift to the processor 39 for each of the second pixel regions 301 to 304. For example Figure 3 in, the reading circuit 31 outputs one first image data I move and four second image data I lift in each frame period. In one embodiment, the first image data I move and the second image data I lift contain the pixel data acquired by the pixel regions during the current exposure period, rather than time-differential pixel data. The frame rates for generating the first image data I move and the second image data I lift can be the same or different from each other, and there is no specific limitation.

[0063] The processor 39 calculates the displacement of the navigation device according to the first image data I move The method of calculating the displacement is known, so its detailed content will not be elaborated here. For example, the processor 39 calculates the displacement of the navigation device by comparing the continuously acquired first image data I move (such as calculating the correlation), and this displacement is used, for example, to control the cursor on the display.

[0064] The processor 39 also determines whether the navigation device is lifted according to the second image data I lift In one implementation, the processor 39 calculates the sum or average pixel value of the second image data I lift The processor 39 then determines whether the navigation device is lifted according to the absolute value of the sum or average pixel value. For example, when the absolute value is greater than or equal to a threshold, it indicates that the navigation device is lifted.

[0065] In another implementation, the processor 39 determines whether the navigation device is lifted according to the absolute value difference between two second pixel regions, such as between 301 and 303 or between 302 and 304. For example, when the difference is greater than or equal to a change threshold, it indicates that the navigation device is lifted; when the difference does not exceed the change threshold, it indicates that the navigation device is not lifted.

[0066] The second average pixel value may be greater than or less than the first average pixel value, depending on whether the second pixel regions 301 to 304 can receive the light emitted by the light source when the navigation device is not lifted. When the processor 39 determines that the time difference (from the first time to the second time) of one of the second pixel regions 301 to 304 is greater than or equal to the change threshold, it indicates that at least one side of the navigation device is lifted by the user. For example, when the absolute value related to the second pixel region 301 is greater than or equal to the change threshold, it indicates that the left side of the navigation device is lifted by the user; when the absolute value related to the second pixel region 302 is greater than or equal to the change threshold, it indicates that the upper side of the navigation device is lifted by the user; and so on.

[0067] When the processor 39 determines that at least one side of the navigation device is lifted by the user, it issues a lift signal S lift for corresponding control, such as stopping controlling the cursor according to the displacement calculated from the first image data I move , entering a power-saving mode to reduce the frame rate of the detection matrix 300 or turning off the light source, etc., but not limited to this. In a non-limiting implementation, the processor 39 performs different controls according to different sides of the lifted navigation device. For example, when it is determined that the upper side of the navigation device is lifted by the user, it controls according to different ratios or multiples based on the first image data I moveThe calculated displacement controls the cursor; and when it is determined that the lower side of the navigation device is lifted by the user, the control of the cursor according to the first image data I is stopped move by the calculated displacement, but is not limited thereto.

[0068] In another embodiment, each pixel of the second pixel regions 301 to 304 is respectively used to output time-difference pixel data to form the second image data I lift . As described above, the time-difference pixel data is the difference in the charges generated by the photodiodes of each pixel at the current time and the reference time; or the difference in the charges generated by the photodiodes of each pixel at the current time and the previous time one frame period ago. In this embodiment, the processor 39 is used to count the number of pixels in each of the second image data I output by the second pixel regions 301 to 304 lift in which the time-difference pixel data of a plurality of pixels is greater than or equal to a predetermined threshold (for example, counting the number of pixels outputting "1"), and based on this, it is determined whether the navigation device is lifted.

[0069] It should be noted that the processor 39 does not necessarily calculate all the pixels of the second image data I lift to determine the lift. Once the count of the pixels with significant time-difference pixel data (i.e., greater than or equal to the predetermined threshold) in a second pixel region is greater than or equal to a predetermined count threshold, the lift is confirmed. After the lift is confirmed, a lift signal S is issued lift and the counting is stopped.

[0070] In other words, when the navigation device is lifted, there will be a large change in the pixel values at different times (i.e., the time-difference pixel data). Therefore, in this embodiment, the processor 31 compares the time-difference pixel data with a predetermined threshold or a predetermined value, and counts the number of pixels with an obvious change (i.e., exceeding the predetermined threshold or value) to determine the lift event. The setting of the predetermined threshold or value can exclude noise interference.

[0071] As described above, when the processor 39 determines that a lift event occurs, predetermined control is performed.

[0072] In an embodiment in which each of the plurality of pixels in the second pixel regions 301 to 304 is used to output time-difference pixel data, the reading circuit 31 also first performs spatial difference on the time-difference pixel data to form a mixed difference. The processor 39 (for example, including a comparator and a counter therein) then determines whether the navigation device is lifted according to the mixed difference.

[0073] More specifically, the reading circuit 31 uses a similar Figures 1A to 1DThe reading frame sequentially reads each of the second pixel regions 301 to 304, and calculates the spatial difference of the time-difference pixel data between the central pixel of the reading frame and the surrounding pixels of the central pixel. Refer to the above description for details. At this time, in order to be able to scan the second pixel regions 301 to 304 sequentially using the reading frame, each of the second pixel regions 301 to 304 includes at least 3×3 pixels. It can be understood that when the reading frame is a pixel region including 5×5 pixels, each of the second pixel regions 301 to 304 includes at least 5×5 pixels.

[0074] Figure 3 In the embodiment shown, each of the second pixel regions 301 to 304 outputs 4 mixed differences, which are obtained by the reading circuit 31 respectively according to 4 reading frames. As described above, the reading circuit 31 calculates the difference between the time-difference pixel data of the central pixel of the reading frame and the time-difference pixel data of each of the surrounding pixels of the central pixel, and calculates the sum of 8 absolute values of the obtained differences as the mixed difference.

[0075] After receiving the mixed difference from the reading circuit 31, the processor 39 compares the mixed difference of each of the reading frames with a predetermined value. When the mixed difference of a certain reading frame is greater than or equal to the predetermined value, the count value is incremented by 1; otherwise, the count value is not increased. The processor 39 accumulates the number of reading frames in which the mixed difference is greater than or equal to the predetermined value for each of the second pixel regions 301 to 304, and when the number of one of the second pixel regions 301 to 304 is greater than or equal to a predetermined number, it is determined that the navigation device is lifted.

[0076] As described above, it is not necessary for the processor 39 to count all the reading frames by scanning all the pixels of each of the second pixel regions 301 to 304. As long as one second pixel region reaches the predetermined value, the lifting of this second pixel region can be confirmed and the scanning can be stopped.

[0077] For example, when the mixed differences of more than 2 (i.e., the threshold is 2) reading frames among the 4 reading frames of the second pixel region 301 (for example, scanning the second pixel region 301 longitudinally, which is determined by the control signal of the timing controller 37) are greater than or equal to the predetermined value, it indicates that the left side of the navigation device is lifted; and so on, the lifting of other sides can be judged. The threshold can be set according to different applications, as long as it is less than or equal to the number of reading frames used for each second pixel region.

[0078] In this embodiment, the processor 39 also calculates the displacement of the navigation device according to the first image data I move Since its content has been described above, it will not be elaborated here.

[0079] As described above, when the processor 39 determines that a lifting event has occurred, predetermined control is performed.

[0080] Figure 3 And Figure 4 The ΔI shown in lift represents the average pixel value change, the statistical pixel number, or the number of cumulative read frames to represent the time change of the charge generated by the photodiode.

[0081] Please refer to Figures 5A to 5C shown, which is a schematic diagram of the storage cabinet 500 with the motion sensor applying the above embodiments of the present invention in different states. For example, Figure 5A shows that at time t1, the cabinet door 51 is open but no object enters the internal space of the storage cabinet 500; Figure 5B shows that at time t2, the cabinet door 51 is open and an object 1000 enters the internal space of the storage cabinet 500 ( Figure 5A And Figure 5B The cross-sectional views of do not show the open cabinet door); Figure 5C shows that at time t3, the cabinet door 51 is closed and an object 1000 exists in the internal space of the storage cabinet 500. Figure 5A And Figure 5B shows that the motion sensor 52 acquires an image while 5C shows that the motion sensor 52 stops acquiring an image.

[0082] The storage cabinet 500 includes a cabinet body 50, a cabinet door 51, and a motion sensor 52. Please refer to Figure 6 In one embodiment, the motion sensor 52 includes a pixel matrix 600, a timing controller 37, a reading circuit 31, and a processor 53. As described above, the timing controller 37 is used to generate control signals so that the reading circuit 31 can read the pixel data of each pixel of the pixel matrix 600, and its detailed description will not be repeated here.

[0083] The cabinet body 50 has an internal space for accommodating an object 1000, such as at least one package, etc. The size and shape of the internal space of the cabinet body 50 are determined according to actual needs and are not particularly limited.

[0084] The cabinet door 51 is used to close / close and open the internal space. In one embodiment, the storage cabinet 500 forms a wireless communication with a mobile device (such as Bluetooth communication, but not limited thereto), so that when the user operates the mobile device to select the cabinet door 51 or the storage cabinet 500, the corresponding cabinet door 51 is opened. In another embodiment, the storage cabinet 500 has a human-machine interface (such as a touch panel or a keyboard), and the cabinet door 51 is opened after the user inputs a password. The cabinet door 51 is locked and unlocked by an electronic lock, for example. When the cabinet door 51 is opened, the motion sensor 52 will receive a signal to know that the cabinet door 51 is opened.

[0085] The motion sensor 52 is disposed inside the storage cabinet 500 and is used to acquire an image of the internal space when the cabinet door 51 is opened. The motion sensor 52 further includes a pixel matrix 600 and a processor 53 (such as an MCU, DSP, or ASIC). As described above, the pixel matrix 600 includes a plurality of pixels (for example, shown as 12×12 pixels, but not limited thereto) arranged in a matrix, and each of the plurality of pixels is used to output time-difference pixel data. As described above, the time-difference pixel data is the charge difference of each of the plurality of pixels at the current time and the reference time, so it will not be elaborated herein.

[0086] The processor 53 is used to determine a plurality of pixels (for example, less than the number of pixels in the pixel matrix 600) that detect motion according to the time-difference pixel data, and determine the object coordinates in the storage cabinet 500 according to the plurality of pixels. For example, Figure 6 In the figure, the diagonal area represents a plurality of pixels that detect motion, and the leading edge of the object 1000 triggers the motion. In order to calculate the object coordinates, in one embodiment, the processor 53 frames a plurality of pixels that detect motion into a pixel region PX_mot, and determines the object coordinates in the storage cabinet 500 based on the pixel region PX_mot.

[0087] For example, Figure 7 Displays a graph of the change of the object coordinates (for example, with the direction of placement in the storage cabinet 500 as the X direction) and time.

[0088] As Figure 7 shown, in the time intervals from approximately 155 seconds to 175 seconds and from approximately 280 seconds to 300 seconds, the X coordinate increases, indicating that the object 1000 enters the storage cabinet 500; in the time intervals from approximately 175 seconds to 195 seconds and from approximately 300 seconds to 345 seconds, the X coordinate decreases, indicating that the object 1000 leaves the storage cabinet 500. It can be understood that Figure 7 This is only an example and is not used to limit the present invention.

[0089] In one embodiment, the processor 53 uses any point, several points averaged, or the centroid of the pixel region PX_mot as the object coordinates, and outputs a direction signal according to the coordinate change of the object coordinates (for example, referring to Figure 7 ), for example, representing the entry direction (for example, Figure 7 when the X coordinate of Figure 7 increases) and the extraction direction (for example, Figure 7 when the X coordinate of Figure 7If the size of the pixel region PX_mot), it is output to the back-end control system to record the object state of the storage cabinet 500.

[0090] In one embodiment, when the processor 53 determines that the charge difference of a certain pixel increases or decreases by more than a predetermined threshold, it is confirmed that the action is detected for the certain pixel. For example Figure 6 Pixels filled with slashes. The processor 53 can then determine the object coordinates and size accordingly.

[0091] In another embodiment, the reading circuit 31 reads the frames (such as Figures 1A to 1D A1 to A36) of the pixel matrix 600 in sequence and calculates the mixed difference of the time-difference pixel data between the central pixel of the read frame and the surrounding pixels of the central pixel. The processor 53 then compares the mixed difference of each of the read frames calculated by the reading circuit 31 with a predetermined value to determine whether an action is detected for a certain pixel. The processor 53 counts the number of read frames in the pixel matrix 600 where the mixed difference is greater than the predetermined value, and generates an activation signal Strig when the number of the read frames is greater than a predetermined number. Since it has been described above, it will not be elaborated here.

[0092] The activation signal Strig is sent to the back-end control system of the storage cabinet 500, indicating that the motion sensor 52 detects the presence of the object 1000 in the internal space of the storage cabinet 500. In one embodiment, after the activation signal Strig is generated, the processor 53 then determines the object coordinates in the storage cabinet 500 based on the time-difference pixel data and records the entry and exit trajectory of the object 1000 accordingly. For example, refer to Figure 7 .

[0093] In this embodiment, the motion sensor 52 only operates after the cabinet door 51 is opened. After the cabinet door 51 is closed / covered, the entry and exit trajectory of the object 1000 is recorded in the back-end control system / platform of the storage cabinet 500.

[0094] The motion sensor 52 is, for example, a CMOS image sensor, which includes a pixel matrix 600 and a processor 53. The pixel matrix 600 includes a plurality of pixels arranged in a matrix for outputting image frames. The processor 53 is used to determine the motion vector of at least one pixel region where the brightness change exceeds the change threshold based on the time-difference image frames, and output a direction signal S_dir accordingly.

[0095] Please refer to Figure 8 As shown, the pixel region where the brightness change exceeds the change threshold, for example, includes a first pixel region PR_d2b where the brightness changes from dark to bright ( Figure 8 The upper right part of which indicates a change from no object at time T0 to having an object at time T1 and Figure 8The lower right part of which represents the change from no object at time T1 to having an object at time T2) and the second pixel region PR_b2d where the brightness changes from bright to dark Figure 8 The upper left part of which represents the change from having an object at time T0 to no object at time T1 and Figure 8 The lower left part of which represents the change from having an object at time T1 to no object at time T2), where Figure 8 It is shown that the object 1000 moves in the X direction from time T1 to T2. If the object 1000 moves in the opposite direction (e.g., towards the left), the first pixel region then becomes Figure 8 The pixel region on the left side (i.e., the brightness changes from dark to bright) and the second pixel region then becomes Figure 8 The pixel region on the right side (i.e., the brightness changes from bright to dark).

[0096] In this embodiment, the processor 53 is used to output the direction signal S_dir according to the motion vector of at least one of the first pixel region PR_d2b and the second pixel region PR_b2d. For example, if the processor 53 calculates the motion vector ΔS1 of the first pixel region PR_d2b of the time difference image frame between time T1 and T0 (displayed as the time interval T1 - T0) and the first pixel region PR_d2b of the time difference image frame between time T2 and T1 (displayed as the time interval T2 - T1) is in the X direction (e.g., Figure 8 The right side direction), the direction signal S_dir is represented by at least one bit as moving to the right. At the same time, the processor 53 also compares whether the motion vector ΔS2 of the second pixel region PR_b2d of the time difference image frame between time T1 and T0 (displayed as the time interval T1 - T0) and the second pixel region PR_b2d of the time difference image frame between time T2 and T1 (displayed as the time interval T2 - T1) has the same direction as ΔS1 for secondary confirmation, so as to improve the judgment accuracy. That is, at least one of ΔS1 and ΔS2 is used to determine the moving direction of the object 1000.

[0097] It can be understood that the time difference image frame is not limited to only containing Figure 8 The two pixel regions shown. When the object 1000 has other features, the time difference image frame contains multiple pixel regions.

[0098] In one implementation, the time difference image frame is obtained by the processor 53 calculating the difference of the image frames output by the pixel matrix 600 at two different times (e.g., T0 and T1 or T1 and T2). In this implementation, the pixel matrix 600 outputs an image frame obtained at one time point rather than the time difference image frame.

[0099] In another implementation, the time difference image frame is directly output by the pixel matrix 600.

[0100] For example, refer to Figure 9A which shows the circuit diagram of a pixel of the pixel matrix 600. The pixel circuit 900A includes a photodiode PD, a transfer transistor SWt, a reset transistor SWrst, a first timing circuit 2a, and a second timing circuit 2b connected to the node V FD .

[0101] The photodiode PD is used to generate light energy according to the received light L. The light energy passes through the transfer transistor SWt and is respectively stored in the first timing circuit 2a and the second timing circuit 2b during different periods (e.g., controlled by the control signal TX). In this embodiment, in addition to storing the light energy during different periods, the first timing circuit 2a and the second timing circuit 2b also convert the light energy into detection signals A and B with corresponding pulse lengths (e.g., T1 and T2) respectively for the subtraction circuit 900C, refer to Figure 9C for subtraction operation. The first timing circuit 2a and the second timing circuit 2b have the same circuit configuration, only the operating periods are different.

[0102] The first timing circuit 2a stores the first light energy (e.g., V Figure 9B shown as T SA ) generated by the photodiode PD during the first period (e.g., refer to Figure 9B shown as V SIG1 ), and outputs the first detection signal A with the first pulse length T1 according to the first light energy V Figure 9B shown as T O1 during the operation period (e.g., refer to SIG1 ).

[0103] The second timing circuit 2b stores the second light energy (e.g., V Figure 9B shown as T SB ) generated by the photodiode PD during the second period (e.g., refer to Figure 9B shown as V SIG2 ), and outputs the second detection signal B with the second pulse length T2 according to the second light energy V O1 during the operation period T SIG2 . It can be understood that the lengths of T1 and T2 in the figure are only examples and are not intended to limit the present invention.

[0104] The subtraction circuit 900C is coupled to the first timing circuit 2a and the second timing circuit 2b, and is used to perform a differential operation on the first detection signal A and the second detection signal B to obtain time-differential pixel data, and all pixels can generate a time-differential image frame. The subtraction circuit 900C has two input terminals respectively coupled to the first timing circuit 2a and the second timing circuit 2b to receive the first detection signal A having a first pulse length T1 and the second detection signal B having a second pulse length T2 respectively. The subtraction circuit 900C includes an operational capacitor Co and a first operational transistor SWA and a second operational transistor SWB connected in series with each other. The operational capacitor Co is connected between the first operational transistor SWA and the second operational transistor SWB. The first operational transistor SWA is used as a switch to control the charging time of the operational capacitor Co by the first current Ic according to the first pulse length T1; the second operational transistor SWB is used as a switch to control the discharging time of the operational capacitor Co by the second current Id according to the second pulse length T2, where the first current Ic is substantially equal to the second current Id. Thereby, the subtraction circuit 900C can perform the numerical calculation of A - B.

[0105] The reset transistor SWrst is coupled between the voltage source V DD and the node V FD for resetting the first timing circuit 2a during the first period T SA and resetting the second timing circuit 2b during the second period T SB

[0106] The transfer transistor SWt is coupled between the photodiode PD and the node V FD for transferring the first optical energy V SA to the first timing circuit 2a for storage during the first period T SIG1 and transferring the second optical energy V SB to the second timing circuit 2b for storage during the second period T SIG2 Thereby, the pixel circuit 900A can be used to store the detected optical energy in different periods to represent the change of the detected light over time.

[0107] In some embodiments, the first timing circuit 2a further includes a first inverter INV1 coupled between the output terminal of the first timing circuit 2a and the subtraction circuit 900C for inverting the first detection signal A; the second timing circuit 2b further includes a second inverter INV2 coupled between the output terminal of the second timing circuit 2b and the subtraction circuit 900C for inverting the second detection signal B. In other embodiments, the first inverter INV1 and the second inverter INV2 are configured in the subtraction circuit 900C instead of being configured in the first timing circuit 2a and the second timing circuit 2b.

[0108] ​For other details of the pixel circuit 900A and the subtraction circuit 900C, reference may be made to U.S. Patent Application No. US18 / 244,296, filed on September 11, 2023, owned by the same assignee as this case, the entire content of which is incorporated herein by reference.

[0109] Please refer to Figures 10A to 10C and Figures 11A to 11C , which are schematic diagrams of the storage cabinet 500 of other embodiments of the present invention in different states (such as shown at times t1, t2, t3). This embodiment also includes an action sensor 52 disposed inside the storage cabinet 500 for obtaining the current image frame of the internal space of the storage cabinet 500. The action sensor 52 also includes a pixel matrix 600 and a processor 53. Refer to Figure 6 . This embodiment further includes a frame buffer 54 for recording the background image frame Im_b. The processor 53 is used to compare the current image frame with the background image frame Im_b to determine whether there is an object 1000 in the internal space.

[0110] It should be noted that although Figures 10A to 10C and Figures 11A to 11C show that the frame buffer 54 is independent of the action sensor 52, it is only for illustration and not for limiting the present invention. In other embodiments, the frame buffer 54 may be included in the action sensor 52.

[0111] Please refer to Figures 10A to 10C As shown, in this embodiment, the background image frame is the image frame obtained by the action sensor 52 when the cabinet door 51 is open and the internal space is empty, for example, Im_b obtained at time t1 is recorded in the frame buffer 54. During operation, the action sensor 52 is used to continuously obtain multiple image frames of the internal space when the cabinet door 51 is open, for example, at time t2; the processor 53 is used to use at least one image frame before the cabinet door 51 is closed as the current image frame and read the background image frame Im_b from the frame buffer 54 for comparison to confirm whether there is an object 1000 in the internal space, wherein the average brightness of the at least one image frame is preferably higher than a predetermined brightness and as close as possible to the time when the cabinet door 51 is closed. That is, when the cabinet door 51 is closed, for example, at time t3, the action sensor 52 does not obtain an image frame, and the processor 53 does not compare the current image frame with the background image frame Im_b. In one embodiment, the storage cabinet 500 further includes a light source (such as a light-emitting diode, but not limited thereto) for lighting when the cabinet door 51 is open to ensure that the current image frame has sufficient brightness.

[0112] Please refer to Figures 11A to 11CAs shown, in this embodiment, the background image frame Im_b is the image frame acquired by the motion sensor 52 when the cabinet door 51 is closed / shut, for example, at time t1; the motion sensor 52 is used to acquire the current image frame of the interior space when the cabinet door 51 is closed / shut (it can acquire only a single image frame), for example, at time t3. Since this embodiment operates when the cabinet door 51 is closed, the storage cabinet 500 preferably further includes a light source 56 (such as a light-emitting diode, but not limited to) disposed inside the storage cabinet 500 for illuminating the interior space after the cabinet door 51 is sequentially opened and closed, that is, signals are generated respectively when the cabinet door 51 is opened and closed. After the motion sensor 52 acquires the current image frame, the light source 56 can be turned off to save power.

[0113] In this embodiment, the light source 56 and the motion sensor 52 preferably stop operating when the cabinet door 51 is opened to save energy consumption, referring to Figure 11B . Figure 11A and Figure 11C in which, when the cabinet door 51 is closed and the motion sensor 52 acquires the image frame, the light source 56 is lit.

[0114] In one embodiment, the frame buffer 54 records multiple background image frames corresponding to storage cabinets 500 of different models (such as different sizes and shapes) before leaving the factory. When the cabinet door 51 is opened, the processor 52 selects the background image frame corresponding to the model of the opened storage cabinet for comparison with the current image frame. Thus, the present invention can be applied to a system with multiple different storage cabinets.

[0115] In one embodiment, a predetermined pattern 50p is disposed (such as drawn, pasted, coated, etc.) on the inner surface of the cabinet body 50 opposite to the motion sensor 52. The predetermined pattern 50p can be, for example, a trademark pattern, a customized pattern, etc. Thereby, the processor 52 can identify the occluded part of the predetermined pattern 50p of the current image frame and the background image frame to determine whether there is an object 1000 in the interior space and the space it occupies.

[0116] The motion sensor of the present invention is, for example, a camera having a processor 53.

[0117] The motion sensor 52 of the present invention can be applied to judge the occupancy status of different spaces, and is not limited to the storage cabinet 500. The processor 53 can judge the occupancy status, congestion status or remaining space of the accommodation space by comparing the background image frame (such as a pre-stored reference image representing no space occupancy) with the current image frame.

[0118] For example, referring to Figure 12 shown, which is a schematic diagram of the motion sensor 52 of the embodiment of the present invention applied to a trash can 120. The motion sensor 52 also includes a frame buffer 54 (referring to Figures 10A - 10C and Figures 11A - 11C) It is used to record the background image frame. For example, when the trash can 120 is empty, the image frame obtained by using the motion sensor 52 is stored in the frame buffer 54 as the background image frame. The processor 53 of the motion sensor 52 judges the occupancy status of the internal space of the trash can 120 by comparing the current image frame (which can be obtained at a predetermined frequency) with the background image frame. For example, when the processor 53 judges that the occupancy status of the internal space exceeds a predetermined ratio, such as 50% to 80%, it issues an activation signal Strig to remind the user to clean the internal space. The activation signal Strig can be sent to the user's mobile device or directly control the indicator of the trash can 120, such as a light or a speaker, to give a warning.

[0119] In one embodiment, the inner surface of the trash can 120 (for example, the inner surface shown within the field of view of the motion sensor 52) is also configured (such as drawn, pasted, coated, etc.) with a predetermined pattern 120p. The processor 53 can then judge the occupancy status of the internal space according to the predetermined pattern 120p (such as the occluded part or the unoccluded part) of the current image frame and the background image frame.

[0120] For example, referring to Figure 13 As shown, it is a schematic diagram of the motion sensor 52 of the embodiment of the present invention applied to the storage space 130 of a tableware recycling area or a storage system. The motion sensor 52 also includes a frame buffer 54 (refer to Figures 10A - 10C and Figures 11A - 11C ) It is used to record the background image frame. For example, when the storage space 130 is empty, the image frame obtained by using the motion sensor 52 is stored in the frame buffer 54 as the background image frame. The processor 53 of the motion sensor 52 judges the occupancy status of the storage space 130 by comparing the current image frame (which can be obtained at a predetermined frequency) with the background image frame. For example, when the processor 53 judges that the occupancy status of the storage space 130 exceeds a predetermined ratio, such as 50% to 80%, it issues an activation signal Strig to remind the user to clean the storage space 130. The activation signal Strig can be sent to the user's mobile device or directly control the indicator of the system, such as a light or a speaker, to give a warning.

[0121] In one embodiment, the inner surface of the storage space 130 (for example, the inner surface shown within the field of view of the motion sensor 52) is also configured (such as drawn, pasted, coated, etc.) with a predetermined pattern 130p. The processor 53 can then judge the occupancy status of the storage space 130 according to the predetermined pattern 130p (such as the occluded part or the unoccluded part) of the current image frame and the background image frame.

[0122] For example, referring to Figure 14As shown, it is a schematic diagram of the motion sensor 52 of the embodiment of the present invention applied to the interior space 140 of a vehicle. The motion sensor 52 also includes a frame buffer 54 (refer to Figures 10A - 10C and Figures 11A - 11C ) for recording background image frames. For example, when the interior space 140 of the vehicle is empty, the image frames obtained by using the motion sensor 52 are stored in the frame buffer 54 as the background image frames. The processor 53 of the motion sensor 52 determines the occupancy status of the interior space of the interior space 140 by comparing the current image frame (obtained at a predetermined frequency) with the background image frame. The processor 53 preferably performs the comparison only when all the vehicle doors are closed. For example, the motion sensor 52 is configured not to obtain the current image frame when at least one vehicle door is open.

[0123] For example, when the engine is turned off or the vehicle interior power is off and all vehicle doors are closed, and the processor 53 determines that there are still people or objects 4000 in the interior space 140 of the vehicle, an activation signal Strig is sent to remind the user to confirm the interior space 140 of the vehicle. The activation signal Strig can be sent to the user's mobile device or directly control the vehicle's indicator, such as a light or a speaker, to give a warning.

[0124] However, the background image frame is not limited to being obtained when the interior space 140 of the vehicle is empty. In another embodiment, the background image frame is obtained by the motion sensor 52 at a predetermined time interval before obtaining the current image frame (even if the interior space 140 is not empty). For example, the background image frame is the previous image frame of the current image frame or a predetermined number of previous image frames. In this embodiment, the background image frame can be updated every time a predetermined time interval elapses.

[0125] In one embodiment, when it is necessary to identify the identity of the people in the vehicle, such as to determine whether it is a baby or a child, the processor 53 of the motion sensor 52 also has a built-in identification model (established by using machine learning algorithms, which is implemented in software and / or hardware). When the engine is turned off or the power is off and it is determined that there are still objects 4000 in the interior space 140 of the vehicle, the processor 53 runs the identification model to distinguish whether there are still people in the vehicle, so as to avoid the situation of accidentally forgetting a baby or a child in the vehicle.

[0126] More specifically, the motion sensor 52 of the present invention can be applied to object identification in any space, and is not limited to the locker, trash can, storage space, and vehicle interior space proposed in the present invention. When necessary, machine learning algorithms and model parameters can also be built into the processor 53 of the motion sensor 52 to perform object recognition and send the recognition results to the backend control system / platform.

[0127] It should be noted that the numerical values in the above embodiments, such as the number of pixels, the number of matrices, and the threshold value, are only examples and are not intended to limit the present invention.

[0128] In summary, in a known sensor with a pixel matrix, at least two frame buffers need to be configured to temporarily store the image frames output by the pixel matrix at different times. Therefore, the present invention further provides a motion sensor (for example, refer to Figures 1A to 1D ) and a navigation device (for example, refer to Figures 3 to 4 ) that use the time-differential pixel data output by the pixel matrix. Then, spatial differentiation is performed on the time-differential pixel data at different positions in the reading frame to complete the hybrid differentiation operation when reading the pixel data, without the need to use a frame buffer. The processor can directly perform subsequent control based on the hybrid differentiation result without performing time-differential operations in the digital stage.

[0129] Although the present invention has been disclosed through the foregoing examples, it is not intended to limit the present invention. Any person with ordinary knowledge and skills in the technical field to which the present invention pertains can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the appended claims.

Claims

1. A storage cabinet, comprising: A cabinet body having an inner space; a cabinet door, the cabinet door being used to close or open the internal space; and A motion sensor is disposed inside the locker and is used to obtain an image of the internal space when the locker door is opened. The motion sensor includes: a pixel matrix, the pixel matrix comprising a plurality of pixels arranged in a matrix, and each of the plurality of pixels is used to output time-differential pixel data; and A processor is used to determine a plurality of pixels that detect motion according to the time-differential pixel data, and to determine the coordinates of the object in the locker according to the plurality of pixels.

2. The storage cabinet according to claim 1, wherein: The time-differential pixel data is a charge difference value generated by each of the plurality of pixels between a current time and a reference time.

3. The storage cabinet according to claim 2, wherein: When the charge difference of a certain pixel increases or decreases beyond a predetermined threshold, the processor determines that the certain pixel detects the action.

4. The storage cabinet according to claim 1, wherein: The processor is further configured to: outputting a direction signal according to the coordinate change of the object coordinates, and The size of the object is determined based on the plurality of pixels.

5. The storage cabinet according to claim 1, further comprising: A reading circuit is used to read the pixel matrix in sequence with a reading frame and calculate the mixed difference of the time-differential pixel data of the central pixel of the reading frame and the surrounding pixels of the central pixel.

6. The storage cabinet according to claim 5, wherein: The processor is further configured to: comparing the mixed difference of each of the read frames calculated by the read circuit with a predetermined value to determine whether a certain pixel detects the action, Counting the number of the read frames in the pixel matrix where the mixed difference is greater than the predetermined value, and An activation signal is generated when the number of the reading frames is greater than a predetermined number.

7. The storage cabinet according to claim 6, wherein: The processor determines the coordinates of the object only after the activation signal is generated.

8. A storage cabinet, comprising: A cabinet body having an inner space; a cabinet door, the cabinet door being used to close or open the internal space; and A motion sensor is disposed inside the locker and is used to obtain an image frame of the internal space when the locker door is opened. The motion sensor includes: a pixel matrix, the pixel matrix comprising a plurality of pixels arranged in a matrix, for outputting the image frame; and A processor is used to determine, based on the time-differential image frame, a motion vector of at least one pixel region whose brightness change exceeds a change threshold, and output a direction signal accordingly. 9 . The locker according to claim 8 , wherein the time-difference image frame is calculated by the processor according to the image frames output by the pixel matrix at two different times.

10. The storage cabinet according to claim 8, wherein: The at least one pixel region includes a first pixel region whose brightness changes from dark to bright and a second pixel region whose brightness changes from bright to dark, and The processor is used for outputting the direction signal according to a motion vector of at least one of the first pixel region and the second pixel region.

11. The storage cabinet according to claim 8, wherein: The time-differenced image frame is directly output by the pixel matrix, and Each of the plurality of pixels comprises: A photodiode for generating light energy; A first timing circuit, the first timing circuit being used for storing the first light energy generated by the photodiode during a first period, and outputting a first detection signal having a first pulse length according to the first light energy during a calculation period; a second timing circuit, the second timing circuit being used for storing the second light energy generated by the photodiode during a second period, and outputting a second detection signal having a second pulse length according to the second light energy during the operation period; and A subtraction circuit is coupled to the first time circuit and the second time circuit and is used for performing a differential operation on the first detection signal and the second detection signal to obtain the time-differential image frame.

12. The locker according to claim 11, wherein: The subtraction circuit comprises: Operational capacitor; a first operational transistor, the first operational transistor being used to control a charging time of the operational capacitor by a first current according to the first pulse length during the operation; and A second operational transistor is used for controlling a discharge time of a second current to the operational capacitor according to the second pulse length during the operation.

13. A storage cabinet, comprising: A cabinet body having an inner space; A cabinet door, which is used to close or open the internal space; A frame buffer, the frame buffer being used for recording background image frames; a motion sensor, the motion sensor being disposed inside the locker and configured to acquire a current image frame of the interior space; and A processor is used to compare the current image frame with the background image frame to determine the object in the internal space.

14. The locker according to claim 13, wherein: The motion sensor is used to continuously acquire multiple image frames of the internal space when the cabinet door is opened, and The processor is used to use at least one image frame of the multiple image frames before the cabinet door is closed as the current image frame.

15. The storage cabinet according to claim 14, wherein: The average brightness of the at least one image frame is higher than a predetermined brightness, and The background image frame is an image frame acquired when the cabinet door is open and the internal space is empty.

16. The storage cabinet according to claim 13, wherein: The background image frame is an image frame acquired by the motion sensor when the cabinet door is closed, and The motion sensor is used to obtain the current image frame of the internal space after the cabinet doors are opened and closed in sequence.

17. The storage cabinet according to claim 16, further comprising: The light source is disposed inside the storage cabinet and is used to illuminate the internal space when the cabinet door is closed and the motion sensor acquires the current image frame.

18. The storage cabinet according to claim 17, wherein: The light source and the motion sensor stop operating when the cabinet door is opened.

19. The storage cabinet according to claim 16, wherein: The frame buffer records a plurality of background image frames corresponding to lockers of different models, and When the cabinet door is opened, the processor selects a background image frame corresponding to the model of the opened cabinet from the frame buffer to compare with the current image frame.

20. The storage cabinet according to claim 13, wherein: The inner surface of the cabinet body opposite to the motion sensor is provided with a predetermined pattern, and The processor is used for comparing the predetermined pattern of the current image frame with the background image frame to determine the object in the interior space.

Images