WO2023280117A1 - 指示信号识别方法、设备以及计算机存储介质 - Google Patents

指示信号识别方法、设备以及计算机存储介质 Download PDF

Info

Publication number
WO2023280117A1
WO2023280117A1 PCT/CN2022/103706 CN2022103706W WO2023280117A1 WO 2023280117 A1 WO2023280117 A1 WO 2023280117A1 CN 2022103706 W CN2022103706 W CN 2022103706W WO 2023280117 A1 WO2023280117 A1 WO 2023280117A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
sequence
indication signal
image
feature sequence
Prior art date
Application number
PCT/CN2022/103706
Other languages
English (en)
French (fr)
Inventor
许云峰
Original Assignee
深圳市道通科技股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 深圳市道通科技股份有限公司 filed Critical 深圳市道通科技股份有限公司
Publication of WO2023280117A1 publication Critical patent/WO2023280117A1/zh

Links

Images

Classifications

    • 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
    • G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/62—Extraction of image or video features relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/08—Feature extraction
    • G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching

Definitions

  • the embodiments of the present invention relate to the technical field of computer data processing, and in particular to an indication signal identification method, device and computer storage medium.
  • embodiments of the present invention provide a method, device, and computer storage medium for identifying indication signals, which are used to solve the problems of identification efficiency and accuracy of indication signals existing in the prior art.
  • an embodiment of the present invention provides a method for identifying an indication signal, the method comprising: acquiring target video data of a target device; performing image recognition on the target video data, and determining target image features corresponding to the target video data Sequence: determining the target indication type of the target device according to the target image feature sequence.
  • the method further includes: determining a plurality of image frames contained in the target video data and timestamps corresponding to each of the image frames; performing image recognition on each of the image frames respectively, The sub-image features corresponding to each of the image frames are obtained; the sub-image features are combined according to the time stamps to obtain the target image feature sequence.
  • the method also includes:
  • each of the optional indicator signal feature sequences corresponds to an optional indicator type; matching the matched optional indicator signal feature sequences to The optional indication type of is determined as the target indication type of the target device.
  • the method also includes:
  • the cycle parameter extracts the image feature sequence to be processed to obtain the target image feature sequence; wherein, the target image feature sequence includes at least one sequence output cycle; the sequence output cycle is the optional indication signal feature sequence output cycle.
  • the sequence period parameter includes a first flag bit and a second flag bit, and the first flag bit and the second flag bit are respectively used to represent the start and end of the sequence output cycle;
  • the method further includes: identifying the first flag bit and the second flag bit in the image feature sequence to be processed; identifying the adjacent first flag bit and the second flag bit.
  • the sequence period parameter also includes a target sequence length; the target sequence length is a sequence length corresponding to the sequence output period; the method further includes: using the target sequence length as The detection step is to detect adjacent and repeated image features in the image feature sequence to be processed; determine the target image feature sequence according to the detection result.
  • the method also includes:
  • the method also includes:
  • the sequence length of the original image feature sequence is The length of the target sequence; the original image feature sequence is matched with each of the optional indicator signal feature sequences; the matched optional indicator signal feature sequence is determined as the target indicator signal feature sequence; the target The indication type corresponding to the indication signal feature sequence is determined as the target indication type.
  • the method further includes: determining the characteristic waveform diagram of the target image according to the characteristic sequence of the target image; the abscissa of the characteristic waveform diagram of the target image is time, and the ordinate is the image characteristic value;
  • the optional indication signal characteristic sequence determines the optional indication signal waveform diagram; the abscissa of the optional indication signal waveform diagram is time, and the vertical coordinate is the indication signal characteristic value; the target image characteristic waveform diagram and the optional Selecting an indication signal waveform diagram to perform waveform matching; determining the optional indication signal characteristic sequence corresponding to the matched optional indication signal waveform diagram as the target indication signal characteristic sequence.
  • an embodiment of the present invention provides an indication signal identification device, including: a processor, a memory, a communication interface, and a communication bus, and the processor, the memory, and the communication interface complete mutual communication through the communication bus. communication among them; the memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform operations such as the indication signal identification method.
  • an embodiment of the present invention provides a computer-readable storage medium, wherein at least one executable instruction is stored in the storage medium, and the executable instruction causes the indication signal identification device to perform the operation of the indication signal identification method .
  • the embodiment of the present invention obtains the target video data of the target device; performs image recognition on the target video data to determine the target image feature sequence corresponding to the target video data; finally determines the target indication type of the target device according to the target image feature sequence, which is different from existing Due to the low efficiency and accuracy of manual identification of indication signals adopted in the technology, the embodiment of the present invention can automatically determine the target image feature sequence according to the target video data, thereby determining the target indication type corresponding to the target image feature sequence, thereby Improve the efficiency and accuracy of indicator signal identification.
  • Fig. 1 shows a schematic flowchart of a method for identifying an indication signal in the prior art.
  • Fig. 2 shows a schematic flowchart of a method for identifying an indication signal provided by an embodiment of the present invention.
  • Fig. 3 shows an application scenario diagram of the indication signal identification method provided by the embodiment of the present invention.
  • Fig. 4 shows a schematic diagram of modules of an indication signal identification system provided by an embodiment of the present invention.
  • Fig. 5 shows a schematic structural diagram of an indication signal identification device provided by an embodiment of the present invention.
  • Fig. 2 shows a flowchart of a method for identifying an indication signal provided by an embodiment of the present invention, and the method is executed by a computer processing device.
  • the computer processing device may include a cell phone, a laptop, and the like.
  • the application scenario of the indication signal identification method can be as shown in Figure 3: the user records or uploads the pre-produced target video data in real time through the diagnostic software installed on the computer processing device, wherein , the shooting object of the target video data is the indicator light on the system under test of the target device.
  • the diagnosis software determines the target indication type corresponding to the target video data through the indication signal identification method provided by the embodiment of the present invention, and returns the target indication type as the recognition result to the user in the diagnosis software.
  • Step 101 Obtain target video data of the target device.
  • the target device may be a device including an indicating device such as a vehicle or a machine, and the indicating device may be a device such as a dashboard or an indicator light.
  • the target video data includes the video data corresponding to the aforementioned pointing device within the target sampling time, multi-continuously shot photo data, and the like.
  • the sampling time may be determined according to the sequence period parameter in step 234 .
  • the target video data may be acquired in real time, or pre-recorded and uploaded by the user.
  • the real-time image feature sequence corresponding to the acquired video data can be determined in real time according to step 102, and whether the real-time image feature sequence satisfies the preset judgment condition can be determined.
  • the preset judgment condition may be that the real-time image feature sequence contains at least one sequence output cycle described in step 234, and the preset judgment condition is used to indicate that the real-time image feature sequence contains all the information required for judging the type of target indication .
  • Step 102 Perform image recognition on the target video data, and determine a target image feature sequence corresponding to the target video data.
  • the process of performing image recognition on the target video data may include the following steps: sampling the target video data according to a certain sampling frequency, and extracting multiple image frames contained in the target video data.
  • Image feature extraction is performed on each image frame, and image feature information such as grayscale information, brightness information, and RGB (Red Green Blue, three primary colors of red, green, and blue) information corresponding to the target area in the image frame is identified. Then, the image feature information corresponding to each target area is combined according to the time sequence in which each image frame appears in the target video data to obtain a target image feature sequence.
  • the target area may be the area where the pointing device is located in each image frame.
  • image preprocessing may be performed on each image frame, thereby improving the efficiency of image feature extraction.
  • step 102 further includes at least: step 1021: determining a plurality of image frames included in the target video data and a time stamp corresponding to each of the image frames.
  • the indication information generally indicates different indication states through the blinking sequence of the indicating device, and the blinking sequence is a sequence including at least one type of sub-signal.
  • the signal characteristic for distinguishing sub-signal types may be the size of the signal value, such as the level, or the duration of the sub-signal, such as long bright and short bright.
  • Each blinking sequence is specifically identified by the type of sub-signals included in the sequence, the timing of the appearance of each sub-signal, the duration of each sub-signal, and the blinking frequency, so that the blinking sequence can be used to characterize various specificities. indication status.
  • each character bit in the blinking sequence represents a unit duration, that is, a unit blinking period, and the duration of a sub-signal includes at least one unit blinking period.
  • the unit blinking period is 1s
  • the indication state of "low oil level” can be represented by the blinking sequence 110101, and 1 means high level , 0 means low level.
  • the actual indication signal output state corresponding to the flashing sequence 110101 is high level 2s-low level 1s-high level 1s-low level 1s-high level 1s.
  • the embodiment of the indicator light is used to describe the flashing sequence, and the duration is longer than or equal to 2 unit flashing periods.
  • the type of sub-signal is determined to be long bright, and the type of sub-signal whose duration is less than 2 and greater than 0 unit flashing periods is determined to be short bright, then 11 means long bright, 1 means short bright, 0 means the indicator light is off, then
  • the actual indication information output state corresponding to the flashing sequence 110101 is long on-off-short on-off-short on.
  • the pointing device has a certain flickering frequency, and the flickering frequency can be used to represent the pointing state, so it is necessary to determine a suitable sampling frequency according to the flickering frequency of the pointing device, and extract multiple image frames from the target video data according to the sampling frequency, so that Avoid missing one or more flicker transformations of the flicker sequence caused by too small sampling frequency, resulting in that the target image feature sequence determined according to the target video data cannot correspond to the indication state, reducing the accuracy of indication information recognition.
  • step 1021 further includes at least: step 211: sampling the target video data according to a preset sampling frequency to obtain the plurality of image frames.
  • the sampling period corresponding to the sampling frequency should be able to cover the minimum time interval of each sub-signal transition in each optional indication signal feature sequence.
  • the optional indication signal characteristic sequence refers to a preset specific indication signal characteristic sequence
  • each optional indication signal characteristic sequence corresponds to an optional indication state
  • each optional indication signal characteristic sequence includes at least one The signal characteristics corresponding to an optional sub-signal.
  • the sampling frequency is 1/2Hz and the total duration of the target video data is 60s
  • an image frame is extracted from the target video data every 2s, and finally 31 image frames are obtained.
  • Step 212 Determine the time stamp corresponding to each image frame according to the time stamp of each image frame in the target video data.
  • the 31 image frames are respectively recorded as P1, P2...P31, and the time stamps of the image frames P1, P2...P31 in the target video data are 00:00 respectively :00, 00:00:02, and 00:00:60.
  • Step 1022 Perform image recognition on each of the image frames to obtain sub-image features corresponding to each of the image frames.
  • the gray distribution information of the image frame is first obtained, and the background area and the target area in the image frame are determined according to the gray distribution information, wherein the target area is the area where the pointing device is located. Then, according to the RGB information and brightness information corresponding to the target area of the image frame, the sub-image features corresponding to each image frame are determined.
  • each image frame may be preprocessed before image recognition, and the image preprocessing may include noise removal, grayscale processing, and binarization processing.
  • the image quality of each image frame before performing image recognition, can also be detected, and when the image quality of the image is not satisfied with the preset threshold, a prompt message is sent to the customer, prompting the user to perform camera parameter adjustment. Adjust or re-upload the target video data, so as to improve the accuracy and efficiency of image recognition.
  • Step 1023 Combine the sub-image features according to the timestamp to obtain the target image feature sequence.
  • All the sub-image features are combined according to the order of the time stamps to obtain a target image feature sequence.
  • the target device in order to improve the delivery rate of the indicator signal, the target device generally sends a blinking sequence several times, so that it can be easily recorded and analyzed by the user. Therefore, in order to improve the efficiency of determining the target indicator type according to the target video data, it is not necessary to The videos are analyzed and matched one by one, and the sequence information corresponding to at least one sequence output cycle is determined to be included in the target image feature sequence.
  • step 1023 further includes at least: step 231 : combining all the sub-image features according to the order of the time stamps to obtain a sequence of image features to be processed.
  • All the sub-image features are combined according to the order of the time stamps to obtain a sequence of image features to be processed.
  • Step 232 Obtain the instruction signal transmission protocol of the target device.
  • the indication signal transmission protocol is used to characterize the transmission and analysis rules of the indication signal, including at least signal flag bits, flashing frequency, on-off contrast, unit signal sequence length, and data analysis protocol.
  • the signal flag bit is used to mark the redundant information related to the transmission state of the indication signal, such as the start of the transmission of the indication signal, the end of the transmission, and whether there is an abnormality in the transmission process;
  • the unit signal sequence length refers to the indication output within a unit blinking period The number of character bits occupied by the signal;
  • the on-off contrast refers to the change value of the brightness of the indicator light when it is on compared to when it is not on.
  • the device identifier of the target device can be obtained, and related device parameters such as the device manufacturer, device model, and communication standard can be determined according to the device identifier, so that the above-mentioned relevant device parameters can be queried in the database to determine the above-mentioned indication signaling protocol.
  • Step 233 Determine a sequence period parameter according to the indication signal transmission protocol.
  • the sequence period parameter refers to parameters related to the output period of the optional indicator signal characteristic sequence, such as the start flag bit information of the period, the sequence length corresponding to each output period, and the like.
  • Step 234 Extract the image feature sequence to be processed according to the sequence period parameter to obtain the target image feature sequence.
  • the image feature sequence to be processed can be identified according to the sequence cycle parameter, and the identified image feature sequence containing at least one sequence output cycle can be extracted as the target image feature sequence, thereby improving the follow-up according to the target.
  • the sequence output period is the output period of the optional indication signal characteristic sequence described in step 212 .
  • the target device first sends the first indication signal, and then sends the second indication signal without interruption after the first indication signal, Assuming that both the first indication signal and the second indication signal contain three sequence output periods respectively, and the target video data is collected at the beginning of the third sequence output period of the first indication signal, then if the target video data is taken to include one sequence output period image feature sequence, it is possible that the target image feature sequence only corresponds to the first indication signal, while the second indication signal after the device state changes is omitted. Therefore, in another embodiment of the present invention, in order to ensure the correctness of the target image feature sequence, the target image feature sequence should include at least two of the aforementioned sequence output periods.
  • the sequence period parameter includes a first flag bit and a second flag bit, and the first flag bit and the second flag bit are respectively used to represent the start and end of the sequence output cycle ; 234 also includes at least: Step 2341: Identify the first flag and the second flag in the image feature sequence to be processed.
  • the first flag bit and the second flag bit may be, for example, S, E, and so on.
  • Step 2342 Determine the image feature sequence to be processed between the identified adjacent first and second flag bits as the target image feature sequence.
  • the image feature sequence to be processed may be 0110S0101010ES0101010ES0101010E0111, and the adjacent first and second flag bits S0101010E are identified, then the target image feature sequence is 0101010.
  • the sequence period parameter also includes a target sequence length; the target sequence length is a set The sequence length corresponding to the sequence output cycle;
  • Step 234 also includes at least:
  • Step 2343 Using the length of the target sequence as the detection step, detect adjacent and repeated image features in the image feature sequence to be processed.
  • the sampling ratio it is also possible to first determine the sampling ratio according to step 313, determine the length of the sampling sequence according to the sampling ratio and the length of the target sequence, and use the length of the sampling sequence as the detection step length to determine the characteristics of the image to be processed.
  • the adjacent and repeated image features are detected in the sequence.
  • the image feature sequence to be processed it is detected sequentially from the first character to find out whether there are image features with a length of 8 character bits and adjacent repeated occurrences.
  • Step 2344 Determine the target image feature sequence according to the detection result.
  • the detected target image feature sequence is A.
  • the image feature sequence to be processed may be 0110010101001010100111.
  • the sampling sequence length may be, for example, 7 characters
  • the target image feature sequence is 0101010.
  • Step 103 Determine the target indication type of the target device according to the target image feature sequence.
  • the target image feature sequence is matched with a plurality of preset optional indicator signal feature sequences, and a table lookup is performed according to the matched optional indicator signal feature sequences to determine the target indicator type.
  • step 103 further includes at least: step 1031 : matching the target image feature sequence with a plurality of optional indication signal feature sequences respectively.
  • each optional indication signal feature sequence corresponds to an optional indication type.
  • step 1031 further includes: Step 311: Obtain the indication signal transmission protocol of the target device.
  • Step 311 is the same as step 232 and will not be repeated here.
  • Step 312 Determine the duration of the output cycle according to the instruction signal transmission protocol; the duration of the output cycle is a duration corresponding to one output cycle of the sequence.
  • the duration of the output cycle may be pre-agreed in the indication signal transmission protocol.
  • the duration of the signal and the length of the target sequence may be determined according to the transmission protocol of the indication signal to determine the duration of the output cycle. Specifically, the product of the signal duration and the target sequence length is determined as the output cycle duration.
  • the isochronous signal refers to a signal indicating that the duration of each sub-signal in the signal output sequence is the same as the interval between sub-signals.
  • Step 313 Determine a sampling ratio according to the sampling frequency and the output cycle duration.
  • the sampling period corresponding to the sampling frequency is determined.
  • the sampling period is the reciprocal of the sampling frequency.
  • Step 314 Perform restoration processing on the target image feature sequence according to the sampling ratio to obtain an original image feature sequence; wherein, the sequence length of the original image feature sequence is the target sequence length.
  • the restoration process refers to merging adjacent and identical character bits of the target image feature sequence at a sampling ratio. For example, when the sampling ratio is 1/3, every 3 adjacent occurrences And the same character bit is merged into one same character bit, for example, after 000 is restored, 0 is obtained.
  • Step 315 Match the original image feature sequence with each of the optional indicator signal feature sequences.
  • Step 316 Determine the matched optional indicator signal feature sequence as the target indicator signal feature sequence.
  • Step 317 Determine the indication type corresponding to the characteristic sequence of the target indication signal as the target indication type.
  • Step 1032 Determine the optional indication type corresponding to the matched optional indication signal feature sequence as the target indication type of the target device.
  • step 1032 further includes: step 321 : determining the characteristic waveform diagram of the target image according to the characteristic sequence of the target image.
  • the characteristic waveform diagram of the target image is plotted with time as the abscissa as time and image feature values as the ordinate.
  • the image feature value may be a character value of each character bit in the target image feature sequence, which may be used to represent, for example, a gray value.
  • Step 322 Determine an optional indication signal waveform diagram according to the optional indication signal feature sequence.
  • the characteristic waveform diagram of the target image is plotted with time as the abscissa as time and the characteristic value of the indication signal as the ordinate.
  • the characteristic value of the indication signal may be a character value of each character bit in the characteristic sequence of the optional indication signal, which may be used to characterize, for example, the level of the signal.
  • Step 323 Perform waveform matching on the characteristic waveform diagram of the target image and the waveform diagram of the optional indication signal.
  • the waveform variation characteristics of the target image characteristic waveform diagram are matched with the waveform variation characteristics of the optional indication signal waveform diagram, wherein the waveform variation characteristics include phase variation frequency, period variation, and amplitude variation.
  • Step 324 Determine the optional indicator signal feature sequence corresponding to the matched optional indicator signal waveform diagram as the target indicator signal feature sequence.
  • the target image characteristic waveform diagram and the optional indication signal waveform diagram have the same waveform change trend, they are deemed to match.
  • the indication signal identification method provided by the embodiment of the present invention obtains the target video data of the target device; performs image recognition on the target video data, determines the target image feature sequence corresponding to the target video data; finally determines the target indication of the target device according to the target image feature sequence Types of. Therefore, different from the problem of low efficiency and accuracy of manual identification of indication signals adopted in the prior art, the indication signal recognition method provided by the embodiment of the present invention can automatically determine the target image feature sequence according to the target video data, thereby determining the The target indication type corresponding to the target image feature sequence, thereby improving the efficiency and accuracy of indication signal identification.
  • Fig. 4 shows a schematic diagram of modules of an indication signal identification system provided by an embodiment of the present invention.
  • the indication signal recognition system 200 is one or more program modules, one or more program modules are stored in memory, and executed by one or more processors to complete the present application, the present application
  • the module referred to in the application refers to a series of computer program instruction segments capable of completing specific functions.
  • the indication signal identification system 200 includes: an acquisition module 201 , an identification module 202 and a determination module 203 .
  • the acquiring module 201 is configured to acquire target video data of the target device.
  • the recognition module 202 is configured to perform image recognition on the target video data, and determine a target image feature sequence corresponding to the target video data.
  • a determining module 203 configured to determine the target indication type of the target device according to the target image feature sequence.
  • the recognition module 202 is further configured to: determine a plurality of image frames contained in the target video data and the time stamps corresponding to each of the image frames; perform image recognition on each of the image frames respectively , to obtain sub-image features corresponding to each of the image frames; combine the sub-image features according to the time stamps to obtain the target image feature sequence.
  • the identification module 202 is further configured to: respectively match the target image feature sequence with multiple optional indicator signal feature sequences; wherein, each of the optional indicator signal feature sequences corresponds to one An optional indication type: determining the optional indication type corresponding to the matched optional indication signal feature sequence as the target indication type of the target device.
  • the determining module 203 is further configured to: respectively match the target image feature sequence with multiple optional indicator signal feature sequences; wherein, each of the optional indicator signal feature sequences corresponds to one An optional indication type: determining the optional indication type corresponding to the matched optional indication signal feature sequence as the target indication type of the target device.
  • the identification module 202 is further configured to: combine all the sub-image features according to the order of the time stamps to obtain a sequence of image features to be processed; obtain the instruction signal transmission protocol of the target device; Determine the sequence period parameter according to the instruction signal transmission protocol; extract the image feature sequence to be processed according to the sequence period parameter to obtain the target image feature sequence; wherein the target image feature sequence includes at least one sequence output Period; the sequence output period is the output period of the optional indication signal characteristic sequence.
  • the sequence period parameter includes a first flag bit and a second flag bit, and the first flag bit and the second flag bit are respectively used to represent the start and end of the sequence output cycle;
  • the recognition module 202 is also used to: identify the first flag and the second flag in the image feature sequence to be processed; identify the adjacent first flag and the second flag The image feature sequence to be processed between flag bits is determined as the target image feature sequence.
  • the sequence period parameter also includes a target sequence length; the target sequence length is a sequence length corresponding to a sequence output period; the identification module 202 is also configured to: use the target sequence length To detect the step size, the adjacent and repeated image features in the image feature sequence to be processed are detected; and the target image feature sequence is determined according to the detection result.
  • the identification module 202 is further configured to: sample the target video data according to the preset sampling frequency to obtain the plurality of image frames; The time stamps in the target video data determine the time stamps corresponding to each of the image frames.
  • the identification module 202 is also used to: obtain the instruction signal transmission protocol of the target device; determine the output cycle duration according to the instruction signal transmission protocol; the output cycle duration is one sequence output The duration corresponding to the cycle; determine the sampling ratio according to the sampling frequency and the output cycle duration; restore the target image feature sequence according to the sampling ratio to obtain the original image feature sequence; wherein, the original image feature sequence The length of the sequence is the length of the target sequence; the original image feature sequence is matched with each of the optional indicator signal feature sequences; the matched optional indicator signal feature sequence is determined as the target indicator signal feature sequence ; Determine the indication type corresponding to the target indication signal feature sequence as the target indication type.
  • the determining module 203 is further configured to: determine the target image characteristic waveform diagram according to the target image characteristic sequence; the abscissa of the target image characteristic waveform diagram is time, and the ordinate is image characteristic value; Determine the optional indication signal waveform diagram according to the optional indication signal characteristic sequence; the abscissa of the optional indication signal waveform diagram is time, and the vertical coordinate is the indication signal characteristic value; combine the target image characteristic waveform diagram with the Perform waveform matching on the waveform diagram of the optional indication signal; determine the characteristic sequence of the optional indication signal corresponding to the matched waveform diagram of the optional indication signal as the characteristic sequence of the target indication signal.
  • the indication signal identification device obtaineds the target video data of the target device; performs image recognition on the target video data, determines the target image feature sequence corresponding to the target video data; finally determines the target indication of the target device according to the target image feature sequence Types of. Therefore, different from the problem of low efficiency and accuracy of manual identification of indication signals adopted in the prior art, the indication signal identification device provided in the embodiment of the present invention can automatically determine the target image feature sequence according to the target video data, thereby determining the The target indication type corresponding to the target image feature sequence, thereby improving the efficiency and accuracy of indication signal recognition.
  • Fig. 5 shows a schematic structural diagram of an indication signal identification device provided by an embodiment of the present invention, and the specific embodiment of the present invention does not limit the specific implementation of the indication signal identification device.
  • the indication signal identification device may include: a processor (processor) 302 , a communication interface (Communications Interface) 304 , a memory (memory) 306 , and a communication bus 308 .
  • the processor 302 , the communication interface 304 , and the memory 306 communicate with each other through the communication bus 308 .
  • the communication interface 304 is configured to communicate with network elements of other devices such as clients or other servers.
  • the processor 302 is configured to execute the indicator signal identification system 200, and specifically, may perform relevant steps in the above embodiments of the indicator signal identification method.
  • the indication signal recognition system 200 may include one or more program modules (refer to FIG. 4 ) composed of program codes, where the program codes include computer-executable instructions.
  • the processor 302 may be a central processing unit CPU, or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention.
  • the one or more processors included in the indication signal identification device may be of the same type, such as one or more CPUs, or may be different types of processors, such as one or more CPUs and one or more ASICs.
  • the memory 306 is used to store the indication signal identification program that constitutes the indication signal identification system 200 .
  • the memory 306 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.
  • the indicator signal identification program that makes up the indicator signal identification system 200 can specifically be invoked by the processor 302 to make the indicator signal identification device perform the following operations: acquire target video data of the target device; perform image recognition on the target video data, and determine the A target image feature sequence corresponding to the target video data; determining a target indication type of the target device according to the target image feature sequence.
  • the indication signal identification program that constitutes the indication signal identification system 200 is invoked by the processor 302 to enable the indication signal identification device to perform the following operations: determine the multiple image frames contained in the target video data and Timestamps corresponding to each of the image frames; performing image recognition on each of the image frames to obtain sub-image features corresponding to each of the image frames; combining the sub-image features according to the time stamps to obtain the Describe the target image feature sequence.
  • the indication signal recognition program constituting the indication signal recognition system 200 is invoked by the processor 302 to make the indication signal identification device perform the following operations: combine the target image feature sequence with a plurality of optional indication signals The characteristic sequences are matched respectively; wherein, each of the optional indication signal characteristic sequences corresponds to an optional indication type; the optional indication type corresponding to the matched optional indication signal characteristic sequence is determined as the target indication of the target device Types of.
  • the indication signal identification program that constitutes the indication signal identification system 200 is invoked by the processor 302 to make the indication signal identification device perform the following operations: perform the following operations according to the order of the time stamps of all the sub-image features Combining to obtain the image feature sequence to be processed; obtaining the instruction signal transmission protocol of the target device; determining the sequence period parameter according to the instruction signal transmission protocol; extracting the image feature sequence to be processed according to the sequence period parameter to obtain The target image feature sequence; wherein, the target image feature sequence includes at least one sequence output period; the sequence output period is the output period of the optional indication signal feature sequence.
  • the sequence period parameter includes a first flag bit and a second flag bit, and the first flag bit and the second flag bit are respectively used to represent the start and end of the sequence output cycle;
  • the indication signal identification program that constitutes the indication signal identification system 200 is invoked by the processor 302 to enable the indication signal identification device to perform the following operations: perform the identification of the first flag bit and the second flag bit in the image feature sequence to be processed Identifying: determining the image feature sequence to be processed between the identified adjacent first flag bits and the second flag bits as the target image feature sequence.
  • the sequence period parameter also includes a target sequence length; the target sequence length is a sequence length corresponding to the sequence output period; the program 310 is invoked by the processor 302 to enable the indication signal to identify
  • the device performs the following operations: using the length of the target sequence as the detection step, to detect adjacent and repeated image features in the image feature sequence to be processed; to determine the target image feature sequence according to the detection result.
  • the indication signal identification program that constitutes the indication signal identification system 200 is invoked by the processor 302 to enable the indication signal identification device to perform the following operations: sample the target video data according to a preset sampling frequency , to obtain the plurality of image frames; and determine the time stamp corresponding to each image frame according to the time stamp of each image frame in the target video data.
  • the indication signal identification program that constitutes the indication signal identification system 200 is invoked by the processor 302 to enable the indication signal identification device to perform the following operations: obtain the indication signal transmission protocol of the target device;
  • the instruction signal transmission protocol determines the duration of the output cycle; the duration of the output cycle is a duration corresponding to the output cycle of the sequence; the sampling ratio is determined according to the sampling frequency and the duration of the output cycle; the target is set according to the sampling ratio
  • the image feature sequence is restored to obtain the original image feature sequence; wherein, the sequence length of the original image feature sequence is the length of the target sequence; the original image feature sequence is respectively compared with each of the optional indication signal feature sequences Matching: determining the matched optional indicator signal feature sequence as a target indicator signal feature sequence; determining an indicator type corresponding to the target indicator signal feature sequence as the target indicator type.
  • the indication signal identification program that constitutes the indication signal identification system 200 is invoked by the processor 302 to enable the indication signal identification device to perform the following operations: determine the target image characteristic waveform diagram according to the target image feature sequence; The abscissa of the target image characteristic waveform diagram is time, and the ordinate is the image feature value; the optional indication signal waveform diagram is determined according to the optional indication signal characteristic sequence; the abscissa of the optional indication signal waveform diagram is time , the ordinate is the characteristic value of the indication signal; performing waveform matching on the characteristic waveform diagram of the target image and the waveform diagram of the optional indication signal; matching the optional indication signal corresponding to the waveform diagram of the optional indication signal The characteristic sequence is determined as the target indicator signal characteristic sequence.
  • the indication signal identification device obtaineds the target video data of the target device; performs image recognition on the target video data, determines the target image feature sequence corresponding to the target video data; finally determines the target indication of the target device according to the target image feature sequence Types of. Therefore, it is different from the problem of low efficiency and accuracy of manual identification of indication signals adopted in the prior art.
  • the indication signal identification device provided by the embodiment of the present invention can automatically determine the target image feature sequence according to the target video data, thereby determining the The target indication type corresponding to the target image feature sequence, thereby improving the efficiency and accuracy of indication signal recognition.
  • An embodiment of the present invention provides a computer-readable storage medium, the storage medium stores at least one executable instruction, and when the executable instruction is run on the indication signal identification device, the indication signal identification device executes any of the above methods The indication signal identification method in the embodiment.
  • the executable instructions may specifically be used to make the indication signal identification device perform the following operations: acquire target video data of the target device; perform image recognition on the target video data, and determine the target image feature sequence corresponding to the target video data; according to the A target image feature sequence determines a target indication type for the target device.
  • the executable instructions cause the indication signal identification device to perform the following operations: determine a plurality of image frames contained in the target video data and a time stamp corresponding to each of the image frames; performing image recognition on each of the image frames to obtain sub-image features corresponding to each of the image frames; combining the sub-image features according to the time stamps to obtain the target image feature sequence.
  • the executable instructions cause the indicator signal recognition device to perform the following operations: respectively match the target image feature sequence with multiple optional indicator signal feature sequences; wherein, each of the The optional indication signal feature sequence corresponds to an optional indication type; and the optional indication type corresponding to the matched optional indication signal feature sequence is determined as the target indication type of the target device.
  • the executable instructions cause the indication signal recognition device to perform the following operations: combine all the sub-image features according to the time stamp order to obtain a sequence of image features to be processed; obtain the The instruction signal transmission protocol of the target device; determine the sequence cycle parameter according to the instruction signal transmission protocol; extract the image feature sequence to be processed according to the sequence cycle parameter to obtain the target image feature sequence; wherein, the The target image characteristic sequence includes at least one sequence output period; the sequence output period is the output period of the optional indication signal characteristic sequence.
  • the sequence period parameter includes a first flag bit and a second flag bit, and the first flag bit and the second flag bit are respectively used to represent the start and end of the sequence output cycle;
  • the executable instructions cause the indication signal identification device to perform the following operations: identify the first flag bit and the second flag bit in the image feature sequence to be processed; The image feature sequence to be processed between the first flag bit and the second flag bit is determined as the target image feature sequence.
  • the sequence cycle parameter also includes a target sequence length; the target sequence length is a sequence length corresponding to one sequence output cycle; the executable instruction causes the indication signal identification device to execute The following operations: using the length of the target sequence as the detection step, detect adjacent and repeated image features in the image feature sequence to be processed; determine the target image feature sequence according to the detection result.
  • the executable instructions cause the indication signal identification device to perform the following operations: sample the target video data according to a preset sampling frequency to obtain the plurality of image frames; The time stamps of the image frames in the target video data determine the time stamps corresponding to each of the image frames.
  • the executable instructions cause the indication signal identification device to perform the following operations: acquire an indication signal transmission protocol of the target device; determine an output cycle duration according to the indication signal transmission protocol; The output cycle duration is a duration corresponding to the sequence output cycle; the sampling ratio is determined according to the sampling frequency and the output cycle duration; the target image feature sequence is restored according to the sampling ratio to obtain the original image feature sequence; wherein, the sequence length of the original image feature sequence is the target sequence length; the original image feature sequence is matched with each of the optional indication signal feature sequences; the matched optional indication The signal feature sequence is determined as the target indication signal feature sequence; and the indication type corresponding to the target indication signal feature sequence is determined as the target indication type.
  • the executable instructions cause the indication signal identification device to perform the following operations: determine the characteristic waveform diagram of the target image according to the characteristic sequence of the target image; the abscissa of the characteristic waveform diagram of the target image is The time and the ordinate are the image characteristic values; the optional indication signal waveform diagram is determined according to the optional indication signal characteristic sequence; the abscissa of the optional indication signal waveform diagram is time, and the ordinate is the indication signal characteristic value; performing waveform matching on the target image characteristic waveform diagram and the optional indication signal waveform diagram; determining the optional indication signal characteristic sequence corresponding to the matched optional indication signal waveform diagram as the target indication signal characteristic sequence .
  • the computer-readable storage medium provided by the embodiment of the present invention obtains the target video data of the target device; performs image recognition on the target video data, determines the target image feature sequence corresponding to the target video data; finally determines the target of the target device according to the target image feature sequence indicates the type. Therefore, it is different from the problem of low efficiency and accuracy of manually identifying indication signals in the prior art.
  • the computer-readable storage medium provided by the embodiment of the present invention can automatically determine the target image feature sequence according to the target video data, thereby determining The target indication type corresponding to the target image feature sequence is obtained, thereby improving the efficiency and accuracy of indication signal recognition.
  • An embodiment of the present invention provides a device for identifying an indication signal, configured to implement the above method for identifying an indication signal.
  • An embodiment of the present invention provides a computer program, and the computer program can be invoked by a processor to enable an indication signal identification device to execute the indication signal identification method in any of the above method embodiments.
  • An embodiment of the present invention provides a computer program product.
  • the computer program product includes a computer program stored on a computer-readable storage medium.
  • the computer program includes program instructions.

Abstract

本发明实施例涉及计算机数据处理技术领域，公开了一种指示信号识别方法，该方法包括：获取目标设备的目标视频数据；对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列；根据所述目标图像特征序列确定所述目标设备的目标指示类型。通过上述方式，本发明实施例提高了指示信号识别的效率。

Description

指示信号识别方法、设备以及计算机存储介质
本申请要求于2021年07月06日提交中国专利局、申请号为202110764237.7、申请名称为“指示信号识别方法、设备以及计算机存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明实施例涉及计算机数据处理技术领域，具体涉及一种指示信号识别方法、设备以及计算机存储介质。
背景技术
目前的控制器或系统上较常通过控制指示灯的灭亮，来生成不同的闪烁序列，从而表征不同的状态信息，把故障、模式等状态信息传递给用户或维修人员。
现有技术中在对闪烁序列进行识别时，一般是如图1所示的，通过人工计时和记录序列，然后在故障表中进行查询，确定指示灯的闪烁序列对应的指示状态。发明人在实施本发明的过程中发现：现有技术中的指示信号的识别效率和准确率都较低。
发明内容
鉴于上述问题，本发明实施例提供了一种指示信号识别方法、设备以及计算机存储介质，用于解决现有技术中存在的指示信号的识别效率和准确率的问题。
一个方面，本发明实施例提供了一种指示信号识别方法，所述方法包括：获取目标设备的目标视频数据；对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列；根据所述目标图像特征序列确定所述目标设备的目标指示类型。
在一种可选的方式中，所述方法还包括：确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳；分别对各个所述图像帧进行图像识别，得到各个所述图像帧分别对应的子图像特征；根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列。
在一种可选的方式中，所述方法还包括：
将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配；其中，每一个所述可选指示信号特征序列对应一个可选指示类型；将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
在一种可选的方式中，所述方法还包括：
根据所述时间戳顺序将所有所述子图像特征进行组合，得到待处理图像特征序列；获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定序列周期参数；根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列；其中，所述目标图像特征序列包括至少一个序列输出周期；所述序列输出周期为所述可选指示信号特征序列的输出周期。
在一种可选的方式中，所述序列周期参数包括第一标志位与第二标志位，所述第一标志位与第二标志位分别用于表征所述序列输出周期的开始与结束；所述方法还包括：在所述待处理图像特征序列进行所述第一标志位和所述第二标志位的识别；将识别到的相邻的所述第一标志位与所述第二标志位之间的所述待处理图像特征序列确定为所述目标图像特征序列。
在一种可选的方式中，所述序列周期参数还包括目标序列长度；所述目标序列长度为一个所述序列输出周期对应的序列长度；所述方法还包括：以所述目标序列长度为检测步长，对所述待处理图像特征序列中进行相邻且重复的图像特征进行检测；根据检测结果确定所述目标图像特征序列。
在一种可选的方式中，所述方法还包括：
按照预设的采样频率对所述目标视频数据进行采样，得到所述多个图像帧；根据各个所述图像帧在所述目标视频数据中的时间戳确定各个所述图像帧对应的时间戳。
在一种可选的方式中，所述方法还包括：
获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定输出周期时长；所述输出周期时长为一个所述序列输出周期对应的时长；
根据所述采样频率与所述输出周期时长确定采样比例；根据所述采样比例对所述目标图像特征序列进行还原处理，得到原始图像特征序列；其中，所述原始图像特征序列的序列长度为所述目标序列长度；将所述原始图像特征序列分别与各个所述可选指示信号特征序列进行匹配；将匹配到的所述可选指示信号特征序列确定为目标指示信号特征序列；将所述目标指示信号特征序列对应的指示类型确定为所述目标指示类型。
在一种可选的方式中，所述方法还包括：根据所述目标图像特征序列确定目标图像特征波形图；所述目标图像特征波形图的横坐标为时间、纵坐标为图像特征值；根据所述可选指示信号特征序列确定可选指示信号波形图；所述可选指示信号波形图的横坐标为时间、纵坐标为指示信号特征值；将所述目标图像特征波形图与所述可选指示信号波形图进行波形匹配；将匹配到的所述可选指示信号波形图对应的所述可选指示信号特征序列确定为所述目标指示信号特征序列。
另一方面，本发明实施例提供了一种指示信号识别设备，包括：处理器、存储器、通信接口和通信总线，所述处理器、所述存储器和所述通信接口通过所述通信总线完成相互间的通信；所述存储器用于存放至少一可执行指令，所述可执行指令使所述处理器执行如所述指示信号识别方法的操作。
又一方面，本发明实施例提供了一种计算机可读存储介质，所述存储介质中存储有至少一可执行指令，所述可执行指令使指示信号识别设备执行所述指示信号识别方法的操作。
本发明实施例通过获取目标设备的目标视频数据；对目标视频数据进行图像识别，确定目标视频数据对应的目标图像特征序列；最后根据目标图像特征序列确定目标设备的目标指示类型，区别于现有技术中采取的人工识别指示信号的效率和准确率都较低的问题，本发明实施例能够自动根据目标视频数据确 定出目标图像特征序列，从而确定出目标图像特征序列对应的目标指示类型，从而提高了指示信号识别的效率和准确率。
上述说明仅是本发明实施例技术方案的概述，为了能够更清楚了解本发明实施例的技术手段，而可依照说明书的内容予以实施，并且为了让本发明实施例的上述和其它目的、特征和优点能够更明显易懂，以下特举本发明的具体实施方式。
附图说明
附图仅用于示出实施方式，而并不认为是对本发明的限制。而且在整个附图中，用相同的参考符号表示相同的部件。
图1示出了现有技术中指示信号识别方法的流程示意图。
图2示出了本发明实施例提供的指示信号识别方法的流程示意图。
图3示出了本发明实施例提供的指示信号识别方法的应用场景图。
图4示出了本发明实施例提供的指示信号识别系统的模块示意图。
图5示出了本发明实施例提供的指示信号识别设备的结构示意图。
具体实施方式
下面将参照附图更详细地描述本发明的示例性实施例。虽然附图中显示了本发明的示例性实施例，然而应当理解，可以以各种形式实现本发明而不应被这里阐述的实施例所限制。
图2示出了本发明实施例提供的指示信号识别方法的流程图，该方法由计算机处理设备执行。该计算机处理设备可以包括手机、笔记本电脑等。
在本发明的一个实施例中，指示信号识别方法的应用场景可以如图3所示出的：用户通过安装在计算机处理设备上的诊断软件，实时录制或者上传预先摄制完成的目标视频数据，其中，目标视频数据的摄制对象是目标设备的被测系统上的指示灯。诊断软件通过本发明实施例示提供的指示信号识别方法确定 出目标视频数据对应的目标指示类型，将该目标指示类型作为识别结果在诊断软件中返回给用户。
下面对本发明实施例提供的指示信号识别方法进行说明，如图2所示，该方法包括以下步骤：步骤101：获取目标设备的目标视频数据。
在本发明的一个实施例中，目标设备可以是如车、机器等包括指示装置的设备，指示装置可以是仪表盘、指示灯等装置。目标视频数据包括前述指示装置在目标采样时间内对应的视频数据、多连拍照片数据等。其中，采样时间可以根据步骤234中的序列周期参数确定。
在本发明的再一个实施例中，目标视频数据可以是实时获取的，也可以是用户预先录制完成并上传的。
当目标视频数据是实时获取的时，可以按照步骤102实时确定获取到的视频数据对应的实时图像特征序列，并确定实时图像特征序列是否满足预设判断条件。在满足预设判断条件时，向用户发送提示消息，以提示结束视频的实时录制。其中，预设判断条件可以是实时图像特征序列中包含至少一个步骤234中所述的序列输出周期，预设判断条件用于表征实时图像特征序列中包含进行目标指示类型的判断所需的全部信息。
步骤102：对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列。
在本发明的一个实施例中，对所述目标视频数据进行图像识别的过程可以包括如下：对目标视频数据按照一定的采样频率进行采样，抽取出目标视频数据中包含的多个图像帧。
对各个图像帧进行图像特征提取，识别出图像帧中的目标区域对应的灰度信息、亮度信息以及RGB(Red Green Blue，红绿蓝三原色)信息等图像特征信息。再按照各个图像帧在目标视频数据中出现的时间顺序将各个目标区域对应的图像特征信息进行组合，得到目标图像特征序列。其中，目标区域可以是各个图像帧中指示装置所在的区域。
在本发明的再一个实施例中，在对各个图像帧进行图像特征提取之前，还可以对各个图像帧进行图像预处理，从而提高图像特征提取的效率。
在本发明的再一个实施例中，步骤102还至少包括：步骤1021：确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳。
首先，指示信息一般是通过指示装置的闪烁序列来表示不同的指示状态，闪烁序列是一段包括至少一种类型的子信号的序列。其中，区分子信号类型的信号特征可以是信号值的大小，如电平的高低，也可以是子信号的持续时长，如长亮与短亮等。
通过序列中包括的子信号的类型、各种子信号出现的时序、各种子信号的持续时长以及闪烁频率等来特异性标识出各个闪烁序列，从而使得闪烁序列可以用于表征各种特异性的指示状态。其中，闪烁序列中的每一字符位代表一个单位持续时长，即单位闪烁周期，一个子信号的持续时长至少包括一个单位闪烁周期。
在本发明的再一个实施例中，当子信号类型为高电平与低电平时，单位闪烁周期为1s，可以用闪烁序列110101表征“油量低”这一指示状态，1表示高电平，0表示低电平。则闪烁序列110101对应于的实际指示信号输出状态为高电平2s-低电平1s-高电平1s-低电平1s-高电平1s。
在本发明的再一个实施例中，当子信号类型为长亮与短亮时，以指示装置为一个指示灯的实施例进行闪烁序列的说明，将持续时长大于或等于2个单位闪烁周期的子信号的种类确定为长亮，将持续时长小于2个并且大于0个单位闪烁周期的子信号的种类确定为短亮，则11表示长亮，1表示短亮，0表示指示灯灭，则闪烁序列110101对应于的实际指示信息输出状态为长亮-灭-短亮-灭-短亮。
指示装置存在一定的闪烁频率，并且闪烁频率可以用于表征指示状态，因此需要根据指示装置的闪烁频率确定适合的采样频率，按照该采样频率从目标视频数据中进行多个图像帧的抽取，从而避免采样频率过小导致的遗漏闪烁序列的某一或多次闪烁变换，从而导致根据目标视频数据确定的目标图像特征序列无法与指示状态对应起来，降低了指示信息识别的准确率。
在本发明的一个实施例中，采样频率可以根据上述各种子信号的持续时长以及各个子信号之间的信号间隔时长来确定。因此，在本发明的再一个实施例 中，步骤1021还至少包括：步骤211：按照预设的采样频率对所述目标视频数据进行采样，得到所述多个图像帧。
在本发明的一个实施例中，采样频率所对应的采样周期应该能够覆盖各个可选指示信号特征序列中每一个子信号变换的最小时间间隔。其中，可选指示信号特征序列指的是预设的特异性的指示信号特征序列，每一个可选指示信号特征序列对应于一种可选指示状态，每一个可选指示信号特征序列包括至少一种可选子信号对应的信号特征。
如在采样频率为1/2Hz，目标视频数据的总时长为60s时，从目标视频数据的起始时间开始，每隔2s从目标视频数据中抽取一个图像帧，则最后得到31个图像帧。
步骤212：根据各个所述图像帧在所述目标视频数据中的时间戳确定各个所述图像帧对应的时间戳。
继续上一步的举例说明，31个图像帧分别记为P1、P2......P31，图像帧P1、P2......P31在目标视频数据中的时间戳分别为00:00:00、00:00:02以及00:00:60。
步骤1022：分别对各个所述图像帧进行图像识别，得到各个所述图像帧分别对应的子图像特征。
在本发明的一个实施例中，首先获取图像帧的灰度分布信息，根据灰度分布信息确定出图像帧中的背景区域以及目标区域，其中目标区域即为指示装置所在的区域。然后根据图像帧目标区域对应的RGB信息以及亮度信息，确定各个图像帧对应的子图像特征。
在本发明的再一个实施例中，在进行图像识别之前可以首先对各个图像帧进行预处理，图像预处理可以包括去噪声处理、灰度化处理以及二值化处理等。
在本发明的再一个实施例中，在进行图像识别之前还可以对各个图像帧的图像质量进行检测，在图像的图像质量不满意预设阈值时，向客户发送提示信息，提示用户进行摄像头参数的调整或者重新上传目标视频数据，从而提高图像识别的准确率和效率。
步骤1023：根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列。
根据所述时间戳的先后顺序将所有所述子图像特征进行组合，得到目标图像特征序列。考虑到目标设备为了提高指示信号的送达率，一个闪烁序列一般会循环数次发送，从而方便被用户记录以及解析，因此，为了提高根据目标视频数据确定目标指示类型的效率，不需要将目标视频一一进行分析和匹配，从确定出包括目标图像特征序列中包括至少一个序列输出周期对应的序列信息进行分析即可。
因此，在本发明的再一个实施例中，步骤1023还至少包括：步骤231：根据所述时间戳顺序将所有所述子图像特征进行组合，得到待处理图像特征序列。
根据所述时间戳的先后顺序将所有所述子图像特征进行组合，得到待处理图像特征序列。
步骤232：获取所述目标设备的指示信号传输协议。
在本发明的一个实施例中，指示信号传输协议用于表征指示信号的传输与解析规则，至少包括如信号标志位、闪烁频率、亮灭对比度、单位信号序列长度以及数据解析协议等。其中，信号标志位用于标志指示信号的传输开始、传输结束、以及传输过程是否出现异常等与指示信号传输状态相关的冗余信息；单位信号序列长度指的是一个单位闪烁周期内输出的指示信号所占的字符位个数；亮灭对比度指的是指示灯在亮起时其亮度相对于不亮时的变化值。
在本发明的再一个实施例中，可以获取目标设备的设备标识，根据设备标识确定设备厂商、设备型号、通信标准等相关设备参数，从而根据上述相关设备参数在数据库中进行查询，确定上述指示信号传输协议。
步骤233：根据所述指示信号传输协议确定序列周期参数。
在本发明的一个实施例中，序列周期参数指的是可选指示信号特征序列的输出周期相关的参数，如周期的起始标志位信息、每一个输出周期对应的序列长度等。
步骤234：根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列。
在本发明的一个实施例中，可以根据序列周期参数对待处理图像特征序列进行识别，将标识到的包含至少一个序列输出周期的图像特征序列抽取出来，作为目标图像特征序列，从而提高后续根据目标图像特征序列进行指示类型判断的效率。其中，所述序列输出周期为步骤212中所述的可选指示信号特征序列的输出周期。
在本发明的再一个实施例中，考虑到可能出现设备状态实时变化的情况，如：目标设备首先发送了第一指示信号，随后在第一指示信号之后不间断地发送了第二指示信号，假设第一指示信号与第二指示信号都分别包含三个序列输出周期，在第一指示信号的第三序列输出周期的开始时刻开始采集目标视频数据，那么若采取包含一个序列输出周期即为目标图像特征序列，则可能目标图像特征序列中对应的只是第一指示信号，而遗漏了设备状态变化后的第二指示信号。因此，在本发明的再一个实施例中，为了确保目标图像特征序列的正确性，目标图像特征序列中应包括至少两个上述序列输出周期。
在本发明的再一个实施例中，所述序列周期参数包括第一标志位与第二标志位，所述第一标志位与第二标志位分别用于表征所述序列输出周期的开始与结束；234还至少包括：步骤2341：在所述待处理图像特征序列进行所述第一标志位和所述第二标志位的识别。
在本发明的一个实施例中，第一标志位和第二标志位可以是如S、E等。
步骤2342：将识别到的相邻的所述第一标志位与所述第二标志位之间的所述待处理图像特征序列确定为所述目标图像特征序列。
举例说明，待处理图像特征序列可以是0110S0101010ES0101010ES0101010E0111，识别出相邻的第一标志位与第二标志位S0101010E，则目标图像特征序列为0101010。在指示信号传输协议中规定了可选指示信号特征序列为定长的情况下，在本发明的再一个实施例中，所述序列周期参数还包括目标序列长度；所述目标序列长度为一个所述序列输出周期对应的序列长度；
步骤234还至少包括：
步骤2343：以所述目标序列长度为检测步长，对所述待处理图像特征序列中进行相邻且重复的图像特征进行检测。
在本发明的再一个实施例中，还可以先按照步骤313确定采样比例，根据采样比例和目标序列长度确定采样序列长度，以采样序列长度为所述检测步长，对所述待处理图像特征序列中进行相邻且重复的图像特征进行检测。其中，将目标序列长度与采样比例的商确定为采样序列长度，如目标序列长度为4，采样比例为1/2，则采样序列长度为4*2＝8。则在待处理图像特在序列中从第一个字符开始顺序进行检测，查找是否存在长度为8个字符位并且相邻重复出现的图像特征。
步骤2344：根据检测结果确定所述目标图像特征序列。
如针对序列BAAC，其中A为长度为7个字符位的序列，检测出的目标图像特征序列即为A。结合步骤2342中的举例，待处理图像特征序列可以是01100101010010101001010100111，在采样序列长度可以是如7个字符位时，则目标图像特征序列为0101010。
步骤103：根据所述目标图像特征序列确定所述目标设备的目标指示类型。
在本发明的一个实施例中，将目标图像特征序列与预设的多个可选指示信号特征序列分别进行匹配，根据匹配到的可选指示信号特征序列进行查表确定目标指示类型。
在本发明的再一个实施例中，步骤103还至少包括：步骤1031：将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配。
其中，每一个所述可选指示信号特征序列对应一个可选指示类型。
在本发明的再一个实施例中，步骤1031还包括：步骤311：获取所述目标设备的指示信号传输协议。
步骤311与步骤232相同，不再赘述。
步骤312：根据所述指示信号传输协议确定输出周期时长；所述输出周期时长为一个所述序列输出周期对应的时长。
在本发明的一个实施例中，输出周期时长可以是预先在指示信号传输协议中约定好的。
在本发明的再一个实施例中，在指示信号为等时信号的情况下，还可以根据指示信号传输协议确定信号持续时长和目标序列长度确定输出周期时长。具体地，将信号持续时长和目标序列长度的乘积确定为输出周期时长。其中，等时信号指的是一个指示信号输出序列中的各个子信号的持续时长和子信号之间的间隔时长是相同的信号。
步骤313：根据所述采样频率和所述输出周期时长确定采样比例。
在本发明的再一个实施例中，首先确定采样频率对应的采样周期。其中，采样周期为采样频率的倒数。将采样周期与一个序列输出周期对应的输出周期时长之间的比值确定为采样比例。举例说明，输出周期时长为5s，采样周期为2s，则采样比例为2/5＝0.4。
步骤314：根据所述采样比例对所述目标图像特征序列进行还原处理，得到原始图像特征序列；其中，所述原始图像特征序列的序列长度为所述目标序列长度。
在本发明的一个实施例中，还原处理指的是以采样比例对目标图像特征序列相邻出现且相同的字符位进行合并，如在采样比例为1/3时，将每3个相邻出现且相同的字符位合并为1个该相同的字符位，如将000进行还原处理后，得到0。
步骤315：将所述原始图像特征序列分别与各个所述可选指示信号特征序列进行匹配。
步骤316：将匹配到的所述可选指示信号特征序列确定为目标指示信号特征序列。
步骤317：将所述目标指示信号特征序列对应的指示类型确定为所述目标指示类型。
步骤1032：将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
在本发明的再一个实施例中，步骤1032还包括：步骤321：根据所述目标图像特征序列确定目标图像特征波形图。
在本发明的一个实施例中，以时间为横坐标为时间、以图像特征值为纵坐标绘制目标图像特征波形图。其中，图像特征值可以是目标图像特征序列中各个字符位上的字符值，可以用于表征如灰度值大小等。
步骤322：根据所述可选指示信号特征序列确定可选指示信号波形图。
在本发明的一个实施例中，以时间为横坐标为时间、以指示信号特征值为纵坐标绘制目标图像特征波形图。其中，指示信号特征值可以是可选指示信号特征序列中各个字符位上的字符值，可以用于表征如电平信号高低等。
步骤323：将所述目标图像特征波形图与所述可选指示信号波形图进行波形匹配。
在本发明的一个实施例中，将目标图像特征波形图的波形变化特征与可选指示信号波形图的波形变化特征进行匹配，其中，波形变化特征包括相位变化频率、周期变化以及振幅变化等。
步骤324：将匹配到的所述可选指示信号波形图对应的所述可选指示信号特征序列确定为所述目标指示信号特征序列。
当目标图像特征波形图与所述可选指示信号波形图的波形变化趋势相同时，视作两者匹配。
本发明实施例提供的指示信号识别方法通过获取目标设备的目标视频数据；对目标视频数据进行图像识别，确定目标视频数据对应的目标图像特征序列；最后根据目标图像特征序列确定目标设备的目标指示类型。从而区别于现有技术中采取的人工识别指示信号的效率和准确率都较低的问题，本发明实施例提供的指示信号识别方法能够自动根据目标视频数据确定出目标图像特征序列，从而确定出目标图像特征序列对应的目标指示类型，从而提高了指示信号识别的效率和准确率。
图4示出了本发明实施例提供的指示信号识别系统的模块示意图。如图4所示，所述指示信号识别系统200为一个或者多个程序模块，一个或者多个程序模块被存储于存储器中，并由一个或多个处理器所执行，以完成本申请，本申请所称的模块是指能够完成特定功能的一系列计算机程序指令段。
所述指示信号识别系统200包括：获取模块201、识别模块202和确定模块203。
在一种可选的方式中，获取模块201，用于获取目标设备的目标视频数据。识别模块202，用于对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列。确定模块203，用于根据所述目标图像特征序列确定所述目标设备的目标指示类型。
在一种可选的方式中，识别模块202还用于：确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳；分别对各个所述图像帧进行图像识别，得到各个所述图像帧分别对应的子图像特征；根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列。
在一种可选的方式中，识别模块202还用于：将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配；其中，每一个所述可选指示信号特征序列对应一个可选指示类型；将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
在一种可选的方式中，确定模块203还用于：将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配；其中，每一个所述可选指示信号特征序列对应一个可选指示类型；将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
在一种可选的方式中，识别模块202还用于：根据所述时间戳顺序将所有所述子图像特征进行组合，得到待处理图像特征序列；获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定序列周期参数；根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列；其中，所述目标图像特征序列包括至少一个序列输出周期；所述序列输出周期为所述可选指示信号特征序列的输出周期。
在一种可选的方式中，所述序列周期参数包括第一标志位与第二标志位，所述第一标志位与第二标志位分别用于表征所述序列输出周期的开始与结束；识别模块202还用于：在所述待处理图像特征序列进行所述第一标志位和所述第二标志位的识别；将识别到的相邻的所述第一标志位与所述第二标志位之间的所述待处理图像特征序列确定为所述目标图像特征序列。
在一种可选的方式中，所述序列周期参数还包括目标序列长度；所述目标序列长度为一个所述序列输出周期对应的序列长度；识别模块202还用于：以所述目标序列长度为检测步长，对所述待处理图像特征序列中进行相邻且重复的图像特征进行检测；根据检测结果确定所述目标图像特征序列。
在一种可选的方式中，识别模块202还用于：按照预设的所述采样频率对所述目标视频数据进行采样，得到所述多个图像帧；根据各个所述图像帧在所述目标视频数据中的时间戳确定各个所述图像帧对应的时间戳。
在一种可选的方式中，识别模块202还用于：获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定输出周期时长；所述输出周期时长为一个所述序列输出周期对应的时长；根据所述采样频率和所述输出周期时长确定采样比例；根据所述采样比例对所述目标图像特征序列进行还原处理，得到原始图像特征序列；其中，所述原始图像特征序列的序列长度为所述目标序列长度；将所述原始图像特征序列分别与各个所述可选指示信号特征序列进行匹配；将匹配到的所述可选指示信号特征序列确定为目标指示信号特征序列；将所述目标指示信号特征序列对应的指示类型确定为所述目标指示类型。
在一种可选的方式中，确定模块203还用于：根据所述目标图像特征序列确定目标图像特征波形图；所述目标图像特征波形图的横坐标为时间、纵坐标为图像特征值；根据所述可选指示信号特征序列确定可选指示信号波形图；所述可选指示信号波形图的横坐标为时间、纵坐标为指示信号特征值；将所述目标图像特征波形图与所述可选指示信号波形图进行波形匹配；将匹配到的所述可选指示信号波形图对应的所述可选指示信号特征序列确定为所述目标指示信号特征序列。
本发明实施例提供的指示信号识别装置通过获取目标设备的目标视频数据；对目标视频数据进行图像识别，确定目标视频数据对应的目标图像特征序列；最后根据目标图像特征序列确定目标设备的目标指示类型。从而区别于现有技术中采取的人工识别指示信号的效率和准确率都较低的问题，本发明实施例提供的指示信号识别装置能够自动根据目标视频数据确定出目标图像特征序列，从而确定出目标图像特征序列对应的目标指示类型，从而提高了指示信号识别的效率和准确率。
图5示出了本发明实施例提供的指示信号识别设备的结构示意图，本发明具体实施例并不对指示信号识别设备的具体实现做限定。
如图5所示，该指示信号识别设备可以包括：处理器(processor)302、通信接口(Communications Interface)304、存储器(memory)306、以及通信总线308。其中：处理器302、通信接口304、以及存储器306通过通信总线308完成相互间的通信。通信接口304，用于与其它设备比如客户端或其它服务器等的网元通信。处理器302，用于执行指示信号识别系统200，具体可以执行上述用于指示信号识别方法实施例中的相关步骤。
具体地，指示信号识别系统200可以包括程序代码组成的一个或多个程序模块(参考图4)，该程序代码包括计算机可执行指令。
处理器302可能是中央处理器CPU，或者是特定集成电路ASIC(Application Specific Integrated Circuit)，或者是被配置成实施本发明实施例的一个或多个集成电路。指示信号识别设备包括的一个或多个处理器，可以是同一类型的处理器，如一个或多个CPU；也可以是不同类型的处理器，如一个或多个CPU以及一个或多个ASIC。
存储器306，用于存放组成指示信号识别系统200的指示信号识别程序。存储器306可能包含高速RAM存储器，也可能还包括非易失性存储器(non-volatile memory)，例如至少一个磁盘存储器。
组成所述指示信号识别系统200的指示信号识别程序具体可以被处理器302调用使指示信号识别设备执行以下操作：获取目标设备的目标视频数据；对所 述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列；根据所述目标图像特征序列确定所述目标设备的目标指示类型。
在一种可选的方式中，组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳；分别对各个所述图像帧进行图像识别，得到各个所述图像帧分别对应的子图像特征；根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列。
在一种可选的方式中，组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配；其中，每一个所述可选指示信号特征序列对应一个可选指示类型；将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
在一种可选的方式中，组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：根据所述时间戳顺序将所有所述子图像特征进行组合，得到待处理图像特征序列；获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定序列周期参数；根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列；其中，所述目标图像特征序列包括至少一个序列输出周期；所述序列输出周期为所述可选指示信号特征序列的输出周期。
在一种可选的方式中，所述序列周期参数包括第一标志位与第二标志位，所述第一标志位与第二标志位分别用于表征所述序列输出周期的开始与结束；组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：在所述待处理图像特征序列进行所述第一标志位和所述第二标志位的识别；将识别到的相邻的所述第一标志位与所述第二标志位之间的所述待处理图像特征序列确定为所述目标图像特征序列。
在一种可选的方式中，所述序列周期参数还包括目标序列长度；所述目标序列长度为一个所述序列输出周期对应的序列长度；所述程序310被处理器302调用使指示信号识别设备执行以下操作：以所述目标序列长度为检测步长，对 所述待处理图像特征序列中进行相邻且重复的图像特征进行检测；根据检测结果确定所述目标图像特征序列。
在一种可选的方式中，组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：按照预设的采样频率对所述目标视频数据进行采样，得到所述多个图像帧；根据各个所述图像帧在所述目标视频数据中的时间戳确定各个所述图像帧对应的时间戳。
在一种可选的方式中，组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定输出周期时长；所述输出周期时长为一个所述序列输出周期对应的时长；根据所述采样频率与和所述输出周期时长确定采样比例；根据所述采样比例对所述目标图像特征序列进行还原处理，得到原始图像特征序列；其中，所述原始图像特征序列的序列长度为所述目标序列长度；将所述原始图像特征序列分别与各个所述可选指示信号特征序列进行匹配；将匹配到的所述可选指示信号特征序列确定为目标指示信号特征序列；将所述目标指示信号特征序列对应的指示类型确定为所述目标指示类型。
在一种可选的方式中，组成所述指示信号识别系统200的指示信号识别程序被处理器302调用使指示信号识别设备执行以下操作：根据所述目标图像特征序列确定目标图像特征波形图；所述目标图像特征波形图的横坐标为时间、纵坐标为图像特征值；根据所述可选指示信号特征序列确定可选指示信号波形图；所述可选指示信号波形图的横坐标为时间、纵坐标为指示信号特征值；将所述目标图像特征波形图与所述可选指示信号波形图进行波形匹配；将匹配到的所述可选指示信号波形图对应的所述可选指示信号特征序列确定为所述目标指示信号特征序列。
本发明实施例提供的指示信号识别设备通过获取目标设备的目标视频数据；对目标视频数据进行图像识别，确定目标视频数据对应的目标图像特征序列；最后根据目标图像特征序列确定目标设备的目标指示类型。从而区别于现有技术中采取的人工识别指示信号的效率和准确率都较低的问题，本发明实施例提供的指示信号识别设备能够自动根据目标视频数据确定出目标图像特征 序列，从而确定出目标图像特征序列对应的目标指示类型，从而提高了指示信号识别的效率和准确率。
本发明实施例提供了一种计算机可读存储介质，所述存储介质存储有至少一可执行指令，该可执行指令在指示信号识别设备上运行时，使得所述指示信号识别设备执行上述任意方法实施例中的指示信号识别方法。
可执行指令具体可以用于使得指示信号识别设备执行以下操作：获取目标设备的目标视频数据；对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列；根据所述目标图像特征序列确定所述目标设备的目标指示类型。
在一种可选的方式中，所述可执行指令使所述指示信号识别设备执行以下操作：确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳；分别对各个所述图像帧进行图像识别，得到各个所述图像帧分别对应的子图像特征；根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列。
在一种可选的方式中，所述可执行指令使所述指示信号识别设备执行以下操作：将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配；其中，每一个所述可选指示信号特征序列对应一个可选指示类型；将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
在一种可选的方式中，所述可执行指令使所述指示信号识别设备执行以下操作：根据所述时间戳顺序将所有所述子图像特征进行组合，得到待处理图像特征序列；获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定序列周期参数；根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列；其中，所述目标图像特征序列包括至少一个序列输出周期；所述序列输出周期为所述可选指示信号特征序列的输出周期。
在一种可选的方式中，所述序列周期参数包括第一标志位与第二标志位，所述第一标志位与第二标志位分别用于表征所述序列输出周期的开始与结束； 所述可执行指令使所述指示信号识别设备执行以下操作：在所述待处理图像特征序列进行所述第一标志位和所述第二标志位的识别；将识别到的相邻的所述第一标志位与所述第二标志位之间的所述待处理图像特征序列确定为所述目标图像特征序列。
在一种可选的方式中，所述序列周期参数还包括目标序列长度；所述目标序列长度为一个所述序列输出周期对应的序列长度；所述可执行指令使所述指示信号识别设备执行以下操作：以所述目标序列长度为检测步长，对所述待处理图像特征序列中进行相邻且重复的图像特征进行检测；根据检测结果确定所述目标图像特征序列。
在一种可选的方式中，所述可执行指令使所述指示信号识别设备执行以下操作：按照预设的采样频率对所述目标视频数据进行采样，得到所述多个图像帧；根据各个所述图像帧在所述目标视频数据中的时间戳确定各个所述图像帧对应的时间戳。
在一种可选的方式中，所述可执行指令使所述指示信号识别设备执行以下操作：获取所述目标设备的指示信号传输协议；根据所述指示信号传输协议确定输出周期时长；所述输出周期时长为一个所述序列输出周期对应的时长；根据所述采样频率与和所述输出周期时长确定采样比例；根据所述采样比例对所述目标图像特征序列进行还原处理，得到原始图像特征序列；其中，所述原始图像特征序列的序列长度为所述目标序列长度；将所述原始图像特征序列分别与各个所述可选指示信号特征序列进行匹配；将匹配到的所述可选指示信号特征序列确定为目标指示信号特征序列；将所述目标指示信号特征序列对应的指示类型确定为所述目标指示类型。
在一种可选的方式中，所述可执行指令使所述指示信号识别设备执行以下操作：根据所述目标图像特征序列确定目标图像特征波形图；所述目标图像特征波形图的横坐标为时间、纵坐标为图像特征值；根据所述可选指示信号特征序列确定可选指示信号波形图；所述可选指示信号波形图的横坐标为时间、纵坐标为指示信号特征值；将所述目标图像特征波形图与所述可选指示信号波形图进行波形匹配；将匹配到的所述可选指示信号波形图对应的所述可选指示信号特征序列确定为所述目标指示信号特征序列。
本发明实施例提供的计算机可读存储介质通过获取目标设备的目标视频数据；对目标视频数据进行图像识别，确定目标视频数据对应的目标图像特征序列；最后根据目标图像特征序列确定目标设备的目标指示类型。从而区别于现有技术中采取的人工识别指示信号的效率和准确率都较低的问题，本发明实施例提供的计算机可读存储介质能够自动根据目标视频数据确定出目标图像特征序列，从而确定出目标图像特征序列对应的目标指示类型，从而提高了指示信号识别的效率和准确率。
本发明实施例提供一种指示信号识别装置，用于执行上述指示信号识别方法。
本发明实施例提供了一种计算机程序，所述计算机程序可被处理器调用使指示信号识别设备执行上述任意方法实施例中的指示信号识别方法。
本发明实施例提供了一种计算机程序产品，计算机程序产品包括存储在计算机可读存储介质上的计算机程序，计算机程序包括程序指令，当程序指令在计算机上运行时，使得所述计算机执行上述任意方法实施例中的指示信号识别方法。
在此提供的算法或显示不与任何特定计算机、虚拟系统或者其它设备固有相关。各种通用系统也可以与基于在此的示教一起使用。根据上面的描述，构造这类系统所要求的结构是显而易见的。此外，本发明实施例也不针对任何特定编程语言。应当明白，可以利用各种编程语言实现在此描述的本发明的内容，并且上面对特定语言所做的描述是为了披露本发明的最佳实施方式。
上述实施例中的步骤，除有特殊说明外，不应理解为对执行顺序的限定。以上仅为本申请的优选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。

Claims (11)

  1. 一种指示信号识别方法，其特征在于，所述方法包括：
    获取目标设备的目标视频数据；
    对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列；
    根据所述目标图像特征序列确定所述目标设备的目标指示类型。
  2. 根据权利要求1所述的方法，其特征在于，所述对所述目标视频数据进行图像识别，确定所述目标视频数据对应的目标图像特征序列，包括：
    确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳；
    分别对各个所述图像帧进行图像识别，得到各个所述图像帧分别对应的子图像特征；
    根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列。
  3. 根据权利要求2所述的方法，其特征在于，所述根据所述目标图像特征序列确定所述目标设备的目标指示类型，包括：
    将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配；其中，每一个所述可选指示信号特征序列对应一个可选指示类型；
    将匹配到的可选指示信号特征序列对应的可选指示类型确定为所述目标设备的目标指示类型。
  4. 根据权利要求3所述的方法，其特征在于，所述根据所述时间戳将所述子图像特征进行组合，得到所述目标图像特征序列，包括：
    根据所述时间戳顺序将所有所述子图像特征进行组合，得到待处理图像特征序列；
    获取所述目标设备的指示信号传输协议；
    根据所述指示信号传输协议确定序列周期参数；
    根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列；其中，所述目标图像特征序列包括至少一个序列输出周期；所述序列输出周期为所述可选指示信号特征序列的输出周期。
  5. 根据权利要求4所述的方法，其特征在于，所述序列周期参数包括第一标志位与第二标志位，所述第一标志位与第二标志位分别用于表征所述序列输出周期的开始与结束；所述根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列，包括：
    在所述待处理图像特征序列进行所述第一标志位和所述第二标志位的识别；
    将识别到的相邻的所述第一标志位与所述第二标志位之间的所述待处理图像特征序列确定为所述目标图像特征序列。
  6. 根据权利要求4所述的方法，其特征在于，所述序列周期参数还包括目标序列长度；所述目标序列长度为一个所述序列输出周期对应的序列长度；所述根据所述序列周期参数对所述待处理图像特征序列进行抽取，得到所述目标图像特征序列，还包括：
    以所述目标序列长度为检测步长，对所述待处理图像特征序列中进行相邻且重复的图像特征进行检测；
    根据检测结果确定所述目标图像特征序列。
  7. 根据权利要求6所述的方法，其特征在于，所述确定所述目标视频数据中包含的多个图像帧和各个所述图像帧对应的时间戳，包括：
    按照预设的采样频率对所述目标视频数据进行采样，得到所述多个图像帧；
    根据各个所述图像帧在所述目标视频数据中的时间戳确定各个所述图像帧对应的时间戳。
  8. 根据权利要求7所述的方法，其特征在于，所述将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配，包括：
    获取所述目标设备的指示信号传输协议；
    根据所述指示信号传输协议确定输出周期时长；所述输出周期时长为一个所述序列输出周期对应的时长；
    根据所述采样频率和所述输出周期时长确定采样比例；
    根据所述采样比例对所述目标图像特征序列进行还原处理，得到原始图像特征序列；其中，所述原始图像特征序列的序列长度为所述目标序列长度；
    将所述原始图像特征序列分别与各个所述可选指示信号特征序列进行匹配；
    将匹配到的所述可选指示信号特征序列确定为目标指示信号特征序列；
    将所述目标指示信号特征序列对应的指示类型确定为所述目标指示类型。
  9. 根据权利要求8所述的方法，其特征在于，所述将所述目标图像特征序列与多个可选指示信号特征序列分别进行匹配，确定所述目标设备的目标指示类型，还包括：
    根据所述目标图像特征序列确定目标图像特征波形图；所述目标图像特征波形图的横坐标为时间、纵坐标为图像特征值；
    根据所述可选指示信号特征序列确定可选指示信号波形图；所述可选指示信号波形图的横坐标为时间、纵坐标为指示信号特征值；
    将所述目标图像特征波形图与所述可选指示信号波形图进行波形匹配；
    将匹配到的所述可选指示信号波形图对应的所述可选指示信号特征序列确定为所述目标指示信号特征序列。
  10. 一种指示信号识别设备，其特征在于，包括：处理器、存储器、通信接口和通信总线，所述处理器、所述存储器和所述通信接口通过所述通信总线完成相互间的通信；
    所述存储器用于存放至少一可执行指令，所述可执行指令使所述处理器执行如权利要求1-9任意一项所述的指示信号识别方法的操作。
  11. 一种计算机可读存储介质，其特征在于，所述存储介质中存储有至少一可执行指令，所述可执行指令被指示信号识别设备的处理器执行时，使得指示信号识别设备执行如权利要求1-9任意一项所述的指示信号识别方法的操作。
PCT/CN2022/103706 2021-07-06 2022-07-04 指示信号识别方法、设备以及计算机存储介质 WO2023280117A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110764237.7 2021-07-06
CN202110764237.7A CN113408475A (zh) 2021-07-06 2021-07-06 指示信号识别方法、设备以及计算机存储介质

Publications (1)

Publication Number Publication Date
WO2023280117A1 true WO2023280117A1 (zh) 2023-01-12

Family

ID=77685320

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/103706 WO2023280117A1 (zh) 2021-07-06 2022-07-04 指示信号识别方法、设备以及计算机存储介质

Country Status (2)

Country Link
CN (1) CN113408475A (zh)
WO (1) WO2023280117A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113408475A (zh) * 2021-07-06 2021-09-17 深圳市道通科技股份有限公司 指示信号识别方法、设备以及计算机存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140270799A1 (en) * 2013-03-13 2014-09-18 Richard D. Roberts Method and system for camera enabled error detection
CN105158684A (zh) * 2015-06-04 2015-12-16 常熟开关制造有限公司(原常熟开关厂） 智能断路器故障显示方法、智能断路器、故障识别终端
CN106855839A (zh) * 2015-12-08 2017-06-16 中国移动通信集团公司 一种服务器故障指示方法及装置
US9818269B1 (en) * 2016-12-16 2017-11-14 Ninad H. Ghodke Status light data transmission
CN108414863A (zh) * 2018-03-13 2018-08-17 北京小米移动软件有限公司 故障诊断方法及装置
CN113408475A (zh) * 2021-07-06 2021-09-17 深圳市道通科技股份有限公司 指示信号识别方法、设备以及计算机存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140270799A1 (en) * 2013-03-13 2014-09-18 Richard D. Roberts Method and system for camera enabled error detection
CN105158684A (zh) * 2015-06-04 2015-12-16 常熟开关制造有限公司(原常熟开关厂） 智能断路器故障显示方法、智能断路器、故障识别终端
CN106855839A (zh) * 2015-12-08 2017-06-16 中国移动通信集团公司 一种服务器故障指示方法及装置
US9818269B1 (en) * 2016-12-16 2017-11-14 Ninad H. Ghodke Status light data transmission
CN108414863A (zh) * 2018-03-13 2018-08-17 北京小米移动软件有限公司 故障诊断方法及装置
CN113408475A (zh) * 2021-07-06 2021-09-17 深圳市道通科技股份有限公司 指示信号识别方法、设备以及计算机存储介质

Also Published As

Publication number Publication date
CN113408475A (zh) 2021-09-17

Similar Documents

Publication Publication Date Title
KR102045871B1 (ko) 인공지능 기술에 기반한 화재감지시스템 및 인공지능 기술에 기반한 화재감지방법
CN107103330A (zh) 一种指示灯状态识别方法和装置
CN107590499B (zh) 一种基于视频的设备led指示灯状态监测方法及系统
CN107209922A (zh) 图像处理设备、图像处理系统、和图像处理方法
WO2023280117A1 (zh) 指示信号识别方法、设备以及计算机存储介质
CN110096945B (zh) 基于机器学习的室内监控视频关键帧实时提取方法
CN111654692B (zh) 摄像头检测方法、装置、摄像头及存储介质
CN112153373A (zh) 明厨亮灶设备的故障识别方法、装置及存储介质
CN107784649A (zh) 基于图像识别的芯棒测试远程报警系统及方法
CN111225299A (zh) 一种onu故障识别、修复方法和装置
CN117372377A (zh) 一种单晶硅棱线的断线检测方法、装置及电子设备
CA2871154A1 (en) System and method of distributed processing for machine-vision analysis
CN115526859A (zh) 生产缺陷识别的方法、分布式处理平台、设备及存储介质
US20220261974A1 (en) Method of detecting an abnormality and abnormality detection device
CN116016909A (zh) 一种电视分区背光检测系统及方法
WO2021243724A1 (zh) 图像处理方法、电子设备、行车记录仪及服务器
CN114973135A (zh) 一种基于头肩的时序视频睡岗识别方法、系统及电子设备
CN111553408B (zh) 视频识别软件自动测试的方法
CN114092965A (zh) 安全帽检测与颜色识别方法、系统、存储介质及计算机设备
CN111988569A (zh) 一种变电站工业视频监控画面着火现象监控方法及系统
CN113923443A (zh) 一种网络录像机测试方法、装置及计算机可读存储介质
CN112073713A (zh) 视频漏录测试方法、装置、设备及存储介质
CN112714284A (zh) 一种电力设备检测方法、装置及移动终端
CN111027466A (zh) 一种银行网点超柜使用规范智能检测方法及装置
CN111402341A (zh) 相机参数确定方法、装置、电子设备及可读存储介质

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE