CN113808117A - Lamp detection method, device, equipment and storage medium - Google Patents
Lamp detection method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN113808117A CN113808117A CN202111122367.7A CN202111122367A CN113808117A CN 113808117 A CN113808117 A CN 113808117A CN 202111122367 A CN202111122367 A CN 202111122367A CN 113808117 A CN113808117 A CN 113808117A
- Authority
- CN
- China
- Prior art keywords
- image
- lamp
- detected
- target frame
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 47
- 238000000034 method Methods 0.000 claims description 30
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 230000007797 corrosion Effects 0.000 claims description 4
- 238000005260 corrosion Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Abstract
The invention relates to a lamp detection method, a device, equipment and a storage medium, wherein the lamp detection method comprises the following steps: acquiring an image to be detected, wherein a scene corresponding to the image to be detected comprises a lamp; carrying out binarization processing on the image to be detected to obtain a binarized image, and determining a lamp candidate frame based on the binarized image; and identifying a candidate area corresponding to the lamp candidate frame in the image to be detected, and determining a lamp target frame of the image to be detected, wherein the lamp target frame is the position of the lamp. Because the lamp candidate frame is screened out through binarization processing, only the candidate area corresponding to the lamp candidate frame needs to be subjected to targeted identification, so that the identification range is narrowed, the identification precision is improved, and the precision and robustness of landscape lamp detection are improved.
Description
Technical Field
The invention relates to the technical field of landscape lamp detection, in particular to a lamp detection method, a device, equipment and a storage medium.
Background
Artificial intelligence technologies represented by computer vision are being used in smart city construction in a large number, wherein city management is a large application field and is characterized by wide, various and scattered demands, such as illegal operation lane occupation detection, garbage detection, smoke and fire detection, landscape lamp detection and the like. In the related art, a deep learning algorithm is used for detecting the landscape lamps, but the landscape lamps of the building are very numerous in shapes, cannot be enumerated, and are few in high-altitude cameras and difficult in data acquisition, so that the effect of completely adopting deep learning to detect the landscape lamps is poor.
Disclosure of Invention
The invention provides a lamp detection method, a device, equipment and a storage medium, which aim to overcome the defects in the related art.
According to a first aspect of the embodiments of the present invention, there is provided a lamp detection method, including:
acquiring an image to be detected, wherein a scene corresponding to the image to be detected comprises a lamp;
carrying out binarization processing on the image to be detected to obtain a binarized image, and determining a lamp candidate frame based on the binarized image;
and identifying a candidate area corresponding to the lamp candidate frame in the image to be detected, and determining a lamp target frame of the image to be detected, wherein the lamp target frame is the position of the lamp.
In one example, the image to be detected is a frame of image of a video to be detected, and a scene corresponding to the video to be detected comprises a lamp;
the method further comprises the following steps:
and determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, wherein the reference image is a frame of image in the video to be detected and before the image to be detected.
In one example, the determining, according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, a light-dark state of a lamp corresponding to the lamp target frame of the reference image includes:
determining the intersection ratio of each lamp target frame of the reference image and each lamp target frame of the image to be detected;
determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is dark when the intersection ratio of the lamp target frame of the reference image and each lamp target frame of the image to be detected is smaller than a preset first proportional threshold;
and under the condition that the intersection ratio of the lamp target frame of the reference image and any lamp target frame of the image to be detected is greater than or equal to the first proportional threshold, determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is bright.
In one example, the determining, according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, a light-dark state of a lamp corresponding to the lamp target frame of the reference image includes:
determining the average value of the pixel values of all pixels in the lamp target frame of the reference image as a first average value, and determining the average value of the pixel values of all pixels in the lamp target frame corresponding to the image to be detected as a second average value;
determining that the light and shade state of the landscape lamp corresponding to the lamp target frame is dark under the condition that the difference value between the first average value and the second average value is greater than a preset first pixel threshold value;
and under the condition that the difference value of the first average value and the second average value is smaller than or equal to a preset first pixel threshold value, determining that the bright-dark state of the landscape lamp corresponding to the lamp target frame is bright.
In one example, the binarizing processing on the image to be detected includes:
determining the pixel difference of each pixel pair between the image to be detected and the reference image, wherein the position of one pixel point in the image to be detected is the same as the position of the other pixel point in the reference image;
determining a first number of pixel pairs having a pixel difference greater than a preset second pixel threshold among all the pixel pairs;
and under the condition that the ratio of the first quantity to the quantity of all pixel pairs is greater than a preset ratio threshold value, carrying out binarization processing on the image to be detected.
In one example, before the binarizing processing is performed on the image to be detected, the method further includes:
and converting the image to be detected into a gray image.
In one example, before determining the luminaire candidate frame based on the binarized image, the method further comprises:
and carrying out image corrosion treatment on the binary image.
In one example, determining a luminaire candidate box based on the binarized image comprises:
and determining a connected region consisting of bright pixels in the binary image as the lamp candidate frame.
In one example, before determining the luminaire candidate frame based on the binarized image, the method further comprises:
and removing the connected region with the area smaller than a preset area threshold value in the binary image.
In one example, the identifying a candidate region in the image to be detected corresponding to the candidate frame of the lamp and determining the target frame of the lamp of the image to be detected includes:
inputting each candidate region into a classification neural network which is trained in advance, and outputting a classification result of each candidate region by the classification neural network, wherein the classification result comprises a lamp and a non-lamp;
and determining the candidate area of the lamp as the classification result as the lamp target frame.
In one example, the acquiring the image to be detected includes:
and acquiring a frame of image with a time interval with the reference image as a preset time length in the video to be detected, and determining the frame of image as the image to be detected.
In one example, further comprising:
and deleting the reference image, saving the image to be detected as the reference image, and saving the lamp target frame of the image to be detected.
According to a second aspect of the embodiments of the present invention, there is provided a luminaire detection apparatus, including:
the device comprises an acquisition module, a detection module and a display module, wherein the acquisition module is used for acquiring an image to be detected, and a scene corresponding to the image to be detected comprises a lamp;
the processing module is used for carrying out binarization processing on the image to be detected to obtain a binarized image and determining a lamp candidate frame based on the binarized image;
and the determining module is used for identifying a candidate area corresponding to the lamp candidate frame in the image to be detected and determining a lamp target frame of the image to be detected, wherein the lamp target frame is the position of the lamp.
In one example, the image to be detected is a frame of image of a video to be detected, and a scene corresponding to the video to be detected comprises a lamp;
the device further comprises a judging module for:
and determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, wherein the reference image is a frame of image in the video to be detected and before the image to be detected.
In one example, the determining module is specifically configured to:
determining the intersection ratio of each lamp target frame of the reference image and each lamp target frame of the image to be detected;
determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is dark when the intersection ratio of the lamp target frame of the reference image and each lamp target frame of the image to be detected is smaller than a preset first proportional threshold;
and under the condition that the intersection ratio of the lamp target frame of the reference image and any lamp target frame of the image to be detected is greater than or equal to the first proportional threshold, determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is bright.
In one example, the determining module is specifically configured to:
determining the average value of the pixel values of all pixels in the lamp target frame of the reference image as a first average value, and determining the average value of the pixel values of all pixels in the lamp target frame corresponding to the image to be detected as a second average value;
determining that the light and shade state of the landscape lamp corresponding to the lamp target frame is dark under the condition that the difference value between the first average value and the second average value is greater than a preset first pixel threshold value;
and under the condition that the difference value of the first average value and the second average value is smaller than or equal to a preset first pixel threshold value, determining that the bright-dark state of the landscape lamp corresponding to the lamp target frame is bright.
In one example, the processing module is specifically configured to:
determining the pixel difference of each pixel pair between the image to be detected and the reference image, wherein the position of one pixel point in the image to be detected is the same as the position of the other pixel point in the reference image;
determining a first number of pixel pairs having a pixel difference greater than a preset second pixel threshold among all the pixel pairs;
and under the condition that the ratio of the first quantity to the quantity of all pixel pairs is greater than a preset ratio threshold value, carrying out binarization processing on the image to be detected.
In one example, the processing module is further to:
and before the binarization processing is carried out on the image to be detected, converting the image to be detected into a gray level image.
In one example, the processing module is further to:
and before the lamp candidate frame is determined based on the binarized image, carrying out image corrosion processing on the binarized image.
In one example, when determining the candidate frame of the luminaire based on the binarized image, the processing module is specifically configured to:
and determining a connected region consisting of bright pixels in the binary image as the lamp candidate frame.
In one example, the processing module is further to:
and removing the connected region with the area smaller than a preset area threshold value in the binary image before determining the lamp candidate frame based on the binary image.
In one example, the determining module is specifically configured to:
inputting each candidate region into a classification neural network which is trained in advance, and outputting a classification result of each candidate region by the classification neural network, wherein the classification result comprises a lamp and a non-lamp;
and determining the candidate area of the lamp as the classification result as the lamp target frame.
In one example, the obtaining module is specifically configured to:
and acquiring a frame of image with a time interval with the reference image as a preset time length in the video to be detected, and determining the frame of image as the image to be detected.
In one example, the system further comprises a caching module configured to:
and deleting the reference image, saving the image to be detected as the reference image, and saving the lamp target frame of the image to be detected.
According to a third aspect of embodiments of the present invention, there is provided an electronic device, the device comprising a memory for storing computer instructions executable on a processor, the processor being configured to implement the method of the first aspect when executing the computer instructions.
According to a fourth aspect of embodiments of the present invention, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of the first aspect.
According to the embodiment, the image to be detected is obtained, binarization processing is carried out on the image to be detected, so that a binarization image is obtained, then the image to be detected is determined as the lamp candidate frame based on the binarization image, and then the candidate area corresponding to the lamp candidate frame in the image to be detected is identified, so that whether each candidate area is a lamp or not is determined, that is, the lamp target frame in the image to be detected is determined, and the lamp target frame is the position of the lamp. Because the lamp candidate frame is screened out through binarization processing, only the candidate area corresponding to the lamp candidate frame needs to be subjected to targeted identification, so that the identification range is narrowed, the identification precision is improved, and the precision and robustness of lamp detection are increased.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a flowchart illustrating a lamp detection method according to an embodiment of the present invention;
FIG. 2 is a flowchart of a lamp detection method according to another embodiment of the present invention
FIG. 3 is a flowchart illustrating a method for determining a lighting fixture dimming state according to an embodiment of the present invention;
FIG. 4 is a flow chart of a method for determining a lamp dimming status according to another embodiment of the present invention;
fig. 5 is a schematic structural diagram of a lamp detection device according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
It is to be understood that although the terms first, second, third, etc. may be used herein to describe various information, these information should not be limited to these terms. These terms are only used to distinguish one type of information from another. For example, first information may also be referred to as second information, and similarly, second information may also be referred to as first information, without departing from the scope of the present invention. The word "if" as used herein may be interpreted as "at … …" or "when … …" or "in response to a determination", depending on the context.
In a first aspect, at least one embodiment of the present invention provides a method for detecting a luminaire, please refer to fig. 1, which illustrates a flow of the method, including steps S101 to S103.
The method may be used for detecting landscape lamps of a building, and may also be used for detecting other types of lamps, and the following mainly takes the landscape lamps as an example to describe the detection method provided in this embodiment, but it should be understood that this is not a limitation on the types of lamps for which the detection method provided in this disclosure is directed.
The landscape lamp is generally in a higher position, so that the target aimed by the detection method can be an image or a video shot by a high-altitude wide-angle camera erected on the roof of a high-rise building, and the high-altitude wide-angle camera can shoot a large range of urban high-rise buildings because of wide visual field. The camera or video shot by the high-altitude wide-angle camera can be shot at night, so that the landscape lamp is obvious in the obtained image or video, and the landscape lamp detected by the method is the lighted landscape lamp or the landscape lamp in a bright state.
In addition, the method may be performed by an electronic device such as a terminal device or a server, where the terminal device may be a User Equipment (UE), a mobile device, a User terminal, a cellular phone, a cordless phone, a Personal Digital Assistant (PDA) handheld device, a computing device, a vehicle-mounted device, a wearable device, or the like, and the method may be implemented by a processor calling computer readable instructions stored in a memory. Alternatively, the method may be performed by a server, which may be a local server, a cloud server, or the like.
In step S101, an image to be detected is obtained, where a scene corresponding to the image to be detected includes a lamp.
The image to be detected can be an image shot by the high-altitude wide-angle camera or a frame of image in a video recorded by the high-altitude wide-angle camera. Since the scene in the shooting view angle of the high-altitude wide-angle camera includes the landscape lamp, the obtained image to be detected includes the image of the landscape lamp. The high-altitude wide-angle camera can shoot images at certain time intervals, and then the images shot each time are obtained to be used as images to be detected; the high-altitude wide-angle camera can continuously shoot videos, video frames can be extracted from the videos according to a certain time interval, and the extracted video frames are used as images to be detected.
In step S102, binarization processing is performed on the image to be detected to obtain a binarized image, and a lamp candidate frame is determined based on the binarized image.
And if the image to be detected is a gray image, directly carrying out binarization processing on the image to be detected. If the image to be detected is a color image, the image to be detected can be converted into a gray image, that is, the corresponding brightness value can be determined according to the red (R), yellow (G) and blue (B) pixel values of each pixel.
During the binarization process, a brightness threshold (e.g. 230) may be set, the brightness of the pixels greater than or equal to the brightness threshold is adjusted to 255, and the brightness of the pixels smaller than the brightness threshold is adjusted to 0, so that the brightness of each pixel is one of 255 (bright pixels) and 0 (dark pixels), a plurality of adjacent bright pixels may form a connected region, a plurality of adjacent dark pixels may form a connected region, and the entire binarized image may include the connected region formed by the bright pixels and the connected region formed by the dark pixels. Since the image to be detected is shot at night, the brightness of the (bright) landscape lamp is higher than that of other objects in the scene, so that the probability that the connected region composed of bright pixels is the landscape lamp is higher, and therefore, the connected region composed of bright pixels in the binary image can be determined as the lamp candidate frame.
After the binarization processing, the binarized image may be subjected to image erosion processing, that is, bright pixels in which all the surrounding pixels (the surrounding range is determined by a preset dimension, for example, a dimension of 2 × 2) are bright pixels are kept high (i.e., the brightness of the bright pixels is kept at 255), and bright pixels in which the surrounding dark pixels are included are adjusted to dark pixels (i.e., the brightness of the bright pixels is adjusted to 0). Through image erosion processing, noise points in the binary image can be eliminated, and the edge of a communication area in the binary image is smoothed.
After the binarization processing, connected regions with areas smaller than a preset area threshold value in the binarized image can be removed, because the connected regions smaller than the area threshold value may be noise points. Since the connected region is determined to be the region where the landscape lamp is located, the area threshold value may be determined according to the lower limit of the area of the region where the landscape lamp is located. By removing the communication area with smaller area, noise points can be reduced, so that the subsequent detection aiming at the candidate area has pertinence, the load is reduced, the efficiency is improved, and the precision is improved.
It is understood that the two noise removing processes (i.e., the image erosion process and the process of removing noise dots with a small area) may be used alternatively or both; when two kinds of processing are used, image erosion processing can be performed first, and then noise points with small areas can be removed.
In step S103, a candidate region corresponding to the candidate frame of the lamp in the image to be detected is identified, and a lamp target frame of the image to be detected is determined, where the lamp target frame is a position where the lamp is located.
Each candidate region can be input to a classification neural network which is trained in advance, and the classification neural network outputs a classification result of each candidate region, wherein the classification result comprises a lamp and a non-lamp; and determining the candidate area of the lamp as the classification result as the lamp target frame.
A training set can be prepared in advance, a plurality of local areas can be selected from sample images in the training set, areas which are images of the lamp in the local areas are marked as a lamp target frame, and the rest local areas are marked as comparison frames; and then inputting the sample image into a classification neural network, outputting a prediction result of each local area, determining a network loss value according to the prediction result and a corresponding labeling result, and further adjusting network parameters of the classification neural network according to the network loss value until the classification neural network converges.
The luminaire target frame may be a smallest rectangular frame that can frame the image of the luminaire. Each lamp target frame is provided with a lamp, and the state of the lamp is a landscape lamp in a bright state, in other words, the lamp target frame of the image to be detected determined in the step is the image of the lamp in the bright state in the image to be detected.
According to the embodiment, the image to be detected is obtained, binarization processing is carried out on the image to be detected, so that a binarization image is obtained, then the lamp candidate frame is determined based on the binarization image, then the candidate area corresponding to the lamp candidate frame in the image to be detected is identified, so as to determine whether each candidate area is a lamp or not, that is, the lamp target frame in the image to be detected is determined, and the lamp target frame is the position of the lamp. Because the lamp candidate frame is screened out through binarization processing, only the candidate area corresponding to the lamp candidate frame needs to be subjected to targeted identification, so that the identification range is narrowed, the identification precision is improved, and the precision and robustness of lamp detection are increased.
In some embodiments of the present disclosure, the image to be detected is a frame image of a video to be detected, and a scene corresponding to the video to be detected includes a lamp;
based on this, referring to fig. 2, after step S101, the method for detecting a luminaire further includes step S104: and determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, wherein the reference image is a frame image in the video to be detected and before the video to be detected.
The detection method provided by this embodiment is repeatedly and circularly executed for each acquired image to be detected, that is, after each acquired image to be detected from the video to be detected, the lamp target frame is determined according to steps S101 to S103. Because the last acquired image to be detected also determines a lamp target frame, the light and shade state of the lamp corresponding to each target frame in the last image to be detected can be determined by comparing the target frame of the last image to be detected with the target frame of the last image to be detected, for example, if a certain target frame in the last image to be detected also has a target frame in the corresponding area of the last image to be detected, the lamp corresponding to the target frame is considered to be in a bright state, namely the lamp is not extinguished between the two times of acquiring the images to be detected; and if a certain target frame in the image to be detected at the last time does not have a target frame in the corresponding area of the image to be detected at this time, the lamp corresponding to the target frame is considered to be in a dark state, namely the lamp is extinguished between two times of obtaining the image to be detected.
It should be noted that, in order to distinguish the images to be detected obtained twice, the image to be detected obtained last time may be referred to as a reference image, and the image to be detected obtained this time may be referred to as an image to be detected continuously. Therefore, the image to be detected and the reference image are two frames of images at different moments in the video to be detected, the reference image is in front, and the image to be detected is behind.
After the lamp target frame is determined by the image to be detected acquired each time, the image to be detected acquired next time needs to be used as a reference image after the lamp target frame is determined by the image to be detected acquired next time, so that the image to be detected acquired each time is stored for standby after the lamp target frame is determined. In addition, because the image to be detected acquired each time is used only once as the reference image, after the light and dark state of the lamp corresponding to each lamp target frame of the reference image is determined in step S104, the reference image can be deleted, the image to be detected is stored as the reference image, and the lamp target frame of the image to be detected is stored, that is, the reference image is updated at any time, so that the memory occupation is reduced, and the determination accuracy of the light and dark state of the lamp is increased.
It can be understood that, after the image to be detected is obtained from the video to be detected for the first time, and the lamp target frame of the image to be detected is determined, step S104 is not executed, but the image to be detected is directly saved as the reference image, and the lamp target frame of the image to be detected is saved.
In an example, the image to be detected may be obtained at a certain time interval, that is, when the image to be detected is obtained, a frame of image with a time interval with the reference image being a preset time length is obtained from the video to be detected, and the frame of image is determined as the image to be detected. The preset time length can be 3s, and the video to be detected can be repeatedly acquired according to the preset time length.
In an example, as shown in fig. 3, the light and shade states of the lamp corresponding to the lamp target frame of the reference image may be determined according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, including steps S301 to S303.
In step S301, an intersection ratio of each luminaire target frame of the reference image and each luminaire target frame of the image to be detected is determined.
And sequentially taking each lamp target frame in the reference image according to the sequence, and sequentially determining the intersection ratio of the lamp target frame and each lamp target frame of the image to be detected after each lamp target frame is obtained.
In step S302, under the condition that the intersection ratio between the lamp target frame of the reference image and each lamp target frame of the to-be-detected image is smaller than a preset first proportional threshold, it is determined that the light-dark state of the lamp corresponding to the lamp target frame of the reference image is dark.
In step S303, when the intersection ratio between the lamp target frame of the reference image and any lamp target frame of the to-be-detected image is greater than or equal to the first ratio threshold, it is determined that the bright-dark state of the lamp corresponding to the lamp target frame of the reference image is bright.
If the intersection ratio of the lamp target frame of the reference image and the lamp target frame of the image to be detected is greater than the preset first proportional threshold, it represents that the scenes corresponding to the two lamp target frames are the same, or correspond to the same lamp in the scene, that is, the lamp target frame in the reference image has a matched lamp target frame in the image to be detected, and it can be preliminarily determined that the lamp corresponding to the lamp target frame is still in a bright state and is not extinguished, so that the bright and dark state of the lamp corresponding to the lamp target frame is bright.
Correspondingly, if the intersection ratio of a certain lamp target frame of the reference image and any lamp target frame of the image to be detected is smaller than a preset ratio threshold, the lamp target frame matched with the lamp target frame in the image to be detected is represented. Therefore, the light and dark states of the lamp corresponding to the lamp target frame of the reference image can be determined to be dark.
In this example, the light and shade state of the lamp is determined by the intersection and comparison of the lamp target frames, and since the lamp target frames are positions where the lamp is located, the determination of the light and shade state of the lamp is simple and accurate.
In another example, as shown in fig. 4, the light and shade states of the lamp corresponding to the lamp target frame of the reference image may be determined according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, including steps S401 to S403.
In step S401, it is determined that the average value of the pixel values of all pixels in the lamp target frame of the reference image is a first average value, and it is determined that the average value of the pixel values of all pixels in the lamp target frame corresponding to the image to be detected is a second average value.
In step S402, when a difference between the first average value and the second average value is greater than a preset first pixel threshold, it is determined that a light-dark state of a luminaire corresponding to the luminaire target frame is dark.
In step S403, when the difference between the first average value and the second average value is smaller than or equal to a preset first pixel threshold, it is determined that the bright-dark state of the lamp corresponding to the frame to be determined by the lamp is bright.
Wherein the first pixel threshold may be preset to 100. And in a dark state, the lamp is characterized to be extinguished after the corresponding moment of the reference image.
In the embodiment, whether the lamp is turned off or not is accurately determined through the average pixel difference of the corresponding target frame, so that the identification precision of the bright and dark state of the lamp is improved.
It is understood that the above-described manner of determining the dimming state of the light fixture in the example shown in fig. 3 and the manner of determining the dimming state of the light fixture in the example shown in fig. 4 may be used alternatively or in combination. When the light and shade state of the lamp is determined by combining the two manners, the manner shown in fig. 3 may be utilized, except that, when step S302 is executed, the lamp target frame of the reference image is determined as a candidate target frame under the condition that the intersection ratio of the lamp target frame of the reference image and each lamp target frame of the image to be detected is smaller than a preset first proportional threshold; then, the method shown in fig. 4 is executed, except that only the average value of the pixel values of all pixels in the candidate target frame is determined as the first average value in step S401, only the average value of the pixel values of all pixels in the lamp target frame corresponding to the candidate target frame in the image to be detected is determined as the second average value, only the first average value and the second average value are used to determine the light and shade state of the lamp corresponding to the lamp candidate frame in steps S402 and S403, that is, when the difference value between the first average value and the second average value is greater than the preset first pixel threshold value, the light and shade state of the lamp corresponding to the lamp candidate frame is determined to be dark, and when the difference value between the first average value and the second average value is less than or equal to the preset first pixel threshold value, the light and shade state of the lamp corresponding to the lamp candidate frame is determined to be bright.
And judging the bright and dark state of the lamp corresponding to the lamp target frame by combining two modes, screening out target ohms which are possibly extinguished by utilizing intersection and comparison between the target frames, namely candidate frames, and then performing pixel comparison on the candidate frames to determine the bright and dark state of the lamp. The accuracy of judging the light and shade state of the lamp is improved, and excessive increase of the operation load is avoided.
In some embodiments of the present disclosure, the image to be detected may be subjected to binarization processing in the following manner: firstly, determining the pixel difference of each pixel pair between the image to be detected and the reference image, wherein the position of one pixel point in the image to be detected in the pixel pair is the same as the position of the other pixel point in the reference image; next, determining a first number of pixel pairs having a pixel difference greater than a preset second pixel threshold among all the pixel pairs; and finally, carrying out binarization processing on the image to be detected under the condition that the ratio of the first quantity to the quantity of all pixel pairs is greater than a preset second ratio threshold value.
The pixel difference may be obtained by subtracting the corresponding two pixels to obtain a difference, and taking an absolute value of the difference. The size of the pixel difference can be used for representing the matching degree of the scene corresponding to the image to be detected and the scene corresponding to the reference image, the pixel difference smaller than or equal to the second pixel threshold value represents that the scenes corresponding to the pixels in the two images are the same, the pixel difference larger than the second pixel threshold value represents that the scenes corresponding to the pixels in the two images are different. Further, the ratio of the first number to the number of all pixel pairs is greater than a preset second ratio threshold, and the scenes corresponding to the two images are represented to be the same, that is, the angles of the high-altitude wide-angle camera when acquiring the two images are the same, and the high-altitude wide-angle camera does not move after acquiring the reference image.
Correspondingly, when the ratio of the first number to the number of all pixel pairs is smaller than or equal to a preset second ratio threshold, it is characterized that the scenes corresponding to the two images are different, that is, the angles of the high-altitude wide-angle camera when acquiring the two images are different, and the high-altitude wide-angle camera moves after acquiring the reference image. Therefore, under the condition, the image to be detected is not subjected to binarization processing, namely the image to be detected is abandoned, and further the image to be detected can be obtained again.
In the embodiment, after the image to be detected is obtained each time, whether the scene corresponding to the image to be detected and the reference image is the same or not is determined, that is, whether the high-altitude wide-angle camera moves or not is determined, if the scene is different, the high-altitude wide-angle camera is abandoned, and if the scene is the same, the light and shade state of the lamp is further determined, so that the problem that the detection result is inaccurate due to the movement of the high-altitude wide-angle camera is solved, and the detection precision of the lamp target frame and the determination precision of the light and shade state of the lamp are improved.
According to a second aspect of the embodiments of the present invention, there is provided a lamp detecting device, please refer to fig. 5, which shows a structure of the device, including:
the acquiring module 501 is configured to acquire an image to be detected, where a scene corresponding to the image to be detected includes a lamp;
a processing module 502, configured to perform binarization processing on the image to be detected to obtain a binarized image, and determine a candidate lamp frame based on the binarized image;
a determining module 503, configured to identify a candidate region in the image to be detected, where the candidate region corresponds to the candidate frame of the lamp, and determine a target frame of the lamp of the image to be detected, where the target frame of the lamp is a location where the lamp is located.
In one example, the image to be detected is a frame of image of a video to be detected, and a scene corresponding to the video to be detected comprises a lamp;
the device further comprises a judging module for:
and determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, wherein the reference image is a frame of image in the video to be detected and before the image to be detected.
In one example, the determining module is specifically configured to:
determining the intersection ratio of each lamp target frame of the reference image and each lamp target frame of the image to be detected;
determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is dark when the intersection ratio of the lamp target frame of the reference image and each lamp target frame of the image to be detected is smaller than a preset first proportional threshold;
and under the condition that the intersection ratio of the lamp target frame of the reference image and any lamp target frame of the image to be detected is greater than or equal to the first proportional threshold, determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is bright.
In one example, the determining module is specifically configured to:
determining the average value of the pixel values of all pixels in the lamp target frame of the reference image as a first average value, and determining the average value of the pixel values of all pixels in the lamp target frame corresponding to the image to be detected as a second average value;
determining that the light and shade state of the landscape lamp corresponding to the lamp target frame is dark under the condition that the difference value between the first average value and the second average value is greater than a preset first pixel threshold value;
and under the condition that the difference value of the first average value and the second average value is smaller than or equal to a preset first pixel threshold value, determining that the bright-dark state of the landscape lamp corresponding to the lamp target frame is bright.
In one example, the processing module is specifically configured to:
determining the pixel difference of each pixel pair between the image to be detected and the reference image, wherein the position of one pixel point in the image to be detected is the same as the position of the other pixel point in the reference image;
determining a first number of pixel pairs having a pixel difference greater than a preset second pixel threshold among all the pixel pairs;
and under the condition that the ratio of the first quantity to the quantity of all pixel pairs is greater than a preset ratio threshold value, carrying out binarization processing on the image to be detected.
In one example, the processing module is further to:
and before the binarization processing is carried out on the image to be detected, converting the image to be detected into a gray level image.
In one example, the processing module is further to:
and before the lamp candidate frame is determined based on the binarized image, carrying out image corrosion processing on the binarized image.
In one example, when determining the candidate frame of the luminaire based on the binarized image, the processing module is specifically configured to:
and determining a connected region consisting of bright pixels in the binary image as the lamp candidate frame.
In one example, the processing module is further to:
and removing the connected region with the area smaller than a preset area threshold value in the binary image before determining the lamp candidate frame based on the binary image.
In one example, the determining module is specifically configured to:
inputting each candidate region into a classification neural network which is trained in advance, and outputting a classification result of each candidate region by the classification neural network, wherein the classification result comprises a lamp and a non-lamp;
and determining the candidate area of the lamp as the classification result as the lamp target frame.
In one example, the obtaining module is specifically configured to:
and acquiring a frame of image with a time interval with the reference image as a preset time length in the video to be detected, and determining the frame of image as the image to be detected.
In one example, the system further comprises a caching module configured to:
and deleting the reference image, saving the image to be detected as the reference image, and saving the lamp target frame of the image to be detected.
With regard to the apparatus in the above-mentioned embodiments, the specific manner in which each module performs the operation has been described in detail in the first aspect with respect to the embodiment of the method, and will not be elaborated here.
In a third aspect, at least one embodiment of the present invention provides an apparatus, please refer to fig. 6, which shows a structure of the apparatus, the apparatus includes a memory for storing computer instructions executable on a processor, and the processor is configured to detect a luminaire based on the method according to any one of the first aspect when executing the computer instructions.
In a fourth aspect, at least one embodiment of the invention provides a computer readable storage medium having a computer program stored thereon, which when executed by a processor, performs the method of any of the first aspects.
In the present invention, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. The term "plurality" means two or more unless expressly limited otherwise.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (15)
1. A lamp detection method is characterized by comprising the following steps:
acquiring an image to be detected, wherein a scene corresponding to the image to be detected comprises a lamp;
carrying out binarization processing on the image to be detected to obtain a binarized image, and determining a lamp candidate frame based on the binarized image;
and identifying a candidate area corresponding to the lamp candidate frame in the image to be detected, and determining a lamp target frame of the image to be detected, wherein the lamp target frame is the position of the lamp.
2. The lamp detection method according to claim 1, wherein the image to be detected is a frame image of a video to be detected, and a scene corresponding to the video to be detected includes a lamp;
the method further comprises the following steps:
and determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image, wherein the reference image is a frame of image in the video to be detected and before the image to be detected.
3. The lamp detection method according to claim 2, wherein the determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image comprises:
determining the intersection ratio of each lamp target frame of the reference image and each lamp target frame of the image to be detected;
determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is dark when the intersection ratio of the lamp target frame of the reference image and each lamp target frame of the image to be detected is smaller than a preset first proportional threshold;
and under the condition that the intersection ratio of the lamp target frame of the reference image and any lamp target frame of the image to be detected is greater than or equal to the first proportional threshold, determining that the light and shade state of the lamp corresponding to the lamp target frame of the reference image is bright.
4. The lamp detection method according to claim 2 or 3, wherein the determining the light and shade state of the lamp corresponding to the lamp target frame of the reference image according to the lamp target frame of the image to be detected and the lamp target frame of the reference image comprises:
determining the average value of the pixel values of all pixels in the lamp target frame of the reference image as a first average value, and determining the average value of the pixel values of all pixels in the lamp target frame corresponding to the image to be detected as a second average value;
determining that the light and shade state of the landscape lamp corresponding to the lamp target frame is dark under the condition that the difference value between the first average value and the second average value is greater than a preset first pixel threshold value;
and under the condition that the difference value of the first average value and the second average value is smaller than or equal to a preset first pixel threshold value, determining that the bright-dark state of the landscape lamp corresponding to the lamp target frame is bright.
5. The lamp detection method according to any one of claims 2 to 4, wherein the binarizing processing of the image to be detected includes:
determining the pixel difference of each pixel pair between the image to be detected and the reference image, wherein the position of one pixel point in the image to be detected is the same as the position of the other pixel point in the reference image;
determining a first number of pixel pairs having a pixel difference greater than a preset second pixel threshold among all the pixel pairs;
and under the condition that the ratio of the first quantity to the quantity of all pixel pairs is greater than a preset ratio threshold value, carrying out binarization processing on the image to be detected.
6. The lamp detection method according to any one of claims 1 to 5, further comprising, before the binarizing processing the image to be detected:
and converting the image to be detected into a gray image.
7. The luminaire detection method according to any one of claims 1 to 6, further comprising, before determining a luminaire candidate frame within the binarized image:
and carrying out image corrosion treatment on the binary image.
8. The luminaire detection method according to any one of claims 1 to 7, wherein determining a luminaire candidate frame based on the binarized image comprises:
and determining a connected region consisting of bright pixels in the binary image as the lamp candidate frame.
9. The luminaire detection method according to claim 8, wherein before determining the luminaire candidate frame based on the binarized image, further comprising:
and removing the connected region with the area smaller than a preset area threshold value in the binary image.
10. The method for detecting a lamp as claimed in any one of claims 1 to 9, wherein the identifying the candidate area corresponding to the candidate frame of the lamp in the image to be detected and determining the target frame of the lamp in the image to be detected comprises:
inputting each candidate region into a classification neural network which is trained in advance, and outputting a classification result of each candidate region by the classification neural network, wherein the classification result comprises a lamp and a non-lamp;
and determining the candidate area of the lamp as the classification result as the lamp target frame.
11. The landscape lamp status method according to any one of claims 2 to 5, wherein the acquiring of the image to be detected comprises:
and acquiring a frame of image with a time interval with the reference image as a preset time length in the video to be detected, and determining the frame of image as the image to be detected.
12. The luminaire detection method of any one of claims 2 to 5, further comprising:
and deleting the reference image, saving the image to be detected as the reference image, and saving the lamp target frame of the image to be detected.
13. A luminaire detection device, comprising:
the device comprises an acquisition module, a detection module and a display module, wherein the acquisition module is used for acquiring an image to be detected, and a scene corresponding to the image to be detected comprises a lamp;
the processing module is used for carrying out binarization processing on the image to be detected to obtain a binarized image and determining a lamp candidate frame based on the binarized image;
and the determining module is used for identifying a candidate area corresponding to the lamp candidate frame in the image to be detected and determining a lamp target frame of the image to be detected, wherein the lamp target frame is the position of the lamp.
14. An electronic device, comprising a memory for storing computer instructions executable on a processor, the processor being configured to implement the method of any one of claims 1 to 12 when executing the computer instructions.
15. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the method of any one of claims 1 to 12.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111122367.7A CN113808117A (en) | 2021-09-24 | 2021-09-24 | Lamp detection method, device, equipment and storage medium |
| PCT/CN2022/119250 WO2023045836A1 (en) | 2021-09-24 | 2022-09-16 | Luminaire detection method and apparatus, device, medium, chip, product, and program |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111122367.7A CN113808117A (en) | 2021-09-24 | 2021-09-24 | Lamp detection method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113808117A true CN113808117A (en) | 2021-12-17 |
Family
ID=78940378
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111122367.7A Pending CN113808117A (en) | 2021-09-24 | 2021-09-24 | Lamp detection method, device, equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113808117A (en) |
| WO (1) | WO2023045836A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023045836A1 (en) * | 2021-09-24 | 2023-03-30 | 上海商汤智能科技有限公司 | Luminaire detection method and apparatus, device, medium, chip, product, and program |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107527511A (en) * | 2016-06-22 | 2017-12-29 | 杭州海康威视数字技术股份有限公司 | A kind of intelligent vehicle driving based reminding method and device |
| US20190147578A1 (en) * | 2017-11-10 | 2019-05-16 | Arcam Ab | Method and apparatus for detecting electron beam source filament wear |
| CN111444975A (en) * | 2020-04-02 | 2020-07-24 | 南通大学 | Traffic light identification method based on image processing and deep learning |
| CN111723860A (en) * | 2020-06-17 | 2020-09-29 | 苏宁云计算有限公司 | Target detection method and device |
| CN111931650A (en) * | 2020-08-11 | 2020-11-13 | 杭州智行星科技有限公司 | Target detection model construction and red light running responsibility tracing method, system, terminal and medium |
| CN112598071A (en) * | 2020-12-28 | 2021-04-02 | 北京市商汤科技开发有限公司 | Open fire identification method, device, equipment and storage medium |
| CN112733818A (en) * | 2021-03-30 | 2021-04-30 | 深圳佑驾创新科技有限公司 | Attention mechanism-based vehicle lamp state identification method, device, terminal and medium |
| US20210192239A1 (en) * | 2019-06-27 | 2021-06-24 | Sensetime Group Limited | Method for recognizing indication information of an indicator light, electronic apparatus and storage medium |
| CN113256560A (en) * | 2021-04-14 | 2021-08-13 | 安徽理工大学 | Heading machine nose area intrusion detection method based on YOLOv5 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112001299B (en) * | 2020-08-21 | 2024-02-02 | 浙江省机电设计研究院有限公司 | Tunnel vehicle finger device and lighting lamp fault identification method |
| CN113808117A (en) * | 2021-09-24 | 2021-12-17 | 北京市商汤科技开发有限公司 | Lamp detection method, device, equipment and storage medium |
-
2021
- 2021-09-24 CN CN202111122367.7A patent/CN113808117A/en active Pending
-
2022
- 2022-09-16 WO PCT/CN2022/119250 patent/WO2023045836A1/en unknown
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107527511A (en) * | 2016-06-22 | 2017-12-29 | 杭州海康威视数字技术股份有限公司 | A kind of intelligent vehicle driving based reminding method and device |
| US20190147578A1 (en) * | 2017-11-10 | 2019-05-16 | Arcam Ab | Method and apparatus for detecting electron beam source filament wear |
| US20210192239A1 (en) * | 2019-06-27 | 2021-06-24 | Sensetime Group Limited | Method for recognizing indication information of an indicator light, electronic apparatus and storage medium |
| CN111444975A (en) * | 2020-04-02 | 2020-07-24 | 南通大学 | Traffic light identification method based on image processing and deep learning |
| CN111723860A (en) * | 2020-06-17 | 2020-09-29 | 苏宁云计算有限公司 | Target detection method and device |
| CN111931650A (en) * | 2020-08-11 | 2020-11-13 | 杭州智行星科技有限公司 | Target detection model construction and red light running responsibility tracing method, system, terminal and medium |
| CN112598071A (en) * | 2020-12-28 | 2021-04-02 | 北京市商汤科技开发有限公司 | Open fire identification method, device, equipment and storage medium |
| CN112733818A (en) * | 2021-03-30 | 2021-04-30 | 深圳佑驾创新科技有限公司 | Attention mechanism-based vehicle lamp state identification method, device, terminal and medium |
| CN113256560A (en) * | 2021-04-14 | 2021-08-13 | 安徽理工大学 | Heading machine nose area intrusion detection method based on YOLOv5 |
Non-Patent Citations (3)
| Title |
|---|
| PETER KAN 等: "DeepLight:light source estimation for augmented reality using deep learning", SPRINGER, 7 May 2019 (2019-05-07), pages 873 - 883, XP036901772, DOI: 10.1007/s00371-019-01666-x * |
| 余震 等: "加权显著图耦合像素决策的红外目标检测算法", 计算机工程与设计, pages 216 - 222 * |
| 贾涛 等: "基于BING和GS的图像目标检测中的快速候选框生成算法", 信息工程大学学报, pages 151 - 157 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023045836A1 (en) * | 2021-09-24 | 2023-03-30 | 上海商汤智能科技有限公司 | Luminaire detection method and apparatus, device, medium, chip, product, and program |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023045836A1 (en) | 2023-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107609470B (en) | Method for detecting early smoke of field fire by video | |
| US8045761B2 (en) | Detection of environmental conditions in a sequence of images | |
| AU2011265429B2 (en) | Method and system for robust scene modelling in an image sequence | |
| CN109726620B (en) | Video flame detection method and device | |
| CN111862195A (en) | Light spot detection method and device, terminal and storage medium | |
| KR101432440B1 (en) | Fire smoke detection method and apparatus | |
| CN110223284B (en) | Detection method and detection device for pinhole camera based on intelligent terminal | |
| US10438072B2 (en) | Video data background tracking and subtraction with multiple layers of stationary foreground and background regions | |
| CN108810423B (en) | Illumination angle adjusting method and system based on image brightness gradient | |
| CN103440484A (en) | Flame detection method adaptive to large outdoor space | |
| CN109413411B (en) | Black screen identification method and device of monitoring line and server | |
| CN111190141B (en) | Method for determining light source emission sequence based on image superposition in visible light indoor positioning | |
| JP7092615B2 (en) | Shadow detector, shadow detection method, shadow detection program, learning device, learning method, and learning program | |
| CN110610485A (en) | Ultra-high voltage transmission line channel hidden danger early warning method based on SSIM algorithm | |
| KR101044903B1 (en) | Fire detecting method using hidden markov models in video surveillance and monitoring system | |
| CN113808117A (en) | Lamp detection method, device, equipment and storage medium | |
| CN109684982B (en) | Flame detection method based on video analysis and combined with miscible target elimination | |
| CN111274896A (en) | Smoke and fire identification algorithm | |
| JP2020088647A (en) | Information processing device, information processing method, and program | |
| JP7092616B2 (en) | Object detection device, object detection method, and object detection program | |
| CN115965934A (en) | Parking space detection method and device | |
| JP7194534B2 (en) | Object detection device, image processing device, object detection method, image processing method, and program | |
| TW201327416A (en) | Method for foreground detection in dynamic background | |
| CN114663299A (en) | Training method and device suitable for image defogging model of underground coal mine | |
| CN110956203B (en) | Static night scene lamplight detection method based on lampunit distribution matching |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40058664 Country of ref document: HK |