CN112149768B - Method for counting cinema audience number by integrating video monitoring and radio frequency identification - Google Patents
Method for counting cinema audience number by integrating video monitoring and radio frequency identification Download PDFInfo
- Publication number
- CN112149768B CN112149768B CN202010991479.5A CN202010991479A CN112149768B CN 112149768 B CN112149768 B CN 112149768B CN 202010991479 A CN202010991479 A CN 202010991479A CN 112149768 B CN112149768 B CN 112149768B
- Authority
- CN
- China
- Prior art keywords
- seat
- cinema
- radio frequency
- audience
- frequency identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012544 monitoring process Methods 0.000 title claims abstract description 26
- 238000013528 artificial neural network Methods 0.000 claims abstract description 13
- 238000003062 neural network model Methods 0.000 claims abstract description 10
- 238000012937 correction Methods 0.000 claims abstract description 5
- 238000012549 training Methods 0.000 claims abstract description 5
- 238000005516 engineering process Methods 0.000 claims description 22
- 238000011176 pooling Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 230000002452 interceptive effect Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 239000002184 metal Substances 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
- G06K17/0029—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device the arrangement being specially adapted for wireless interrogation of grouped or bundled articles tagged with wireless record carriers
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/183—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
Abstract
The invention relates to a method for counting cinema audience numbers by integrating video monitoring and radio frequency identification, which comprises the following steps: under the condition of normal seating, a computer controls a camera to shoot an image, each seat in the image is calibrated, then the image of each seat is intercepted according to the coordinates of each seat in the image, and reference correction is carried out; the method comprises the steps of inputting seat images with audience or not into a deep neural network for training to obtain a deep neural network model for judging whether the seats are people or not, enabling a reader-writer to identify all the tags of the seats in a cinema, counting the signal intensity range of radio frequency identification tags of each seat, and establishing a signal intensity model; at the appointed moment in the projection process, the computer controls the camera to shoot images, the images are input into the deep neural network model, whether audience seats exist or not is accurately identified, and the computer calculates the audience numbers according to the label numbers obtained through reference identification and the label numbers obtained through multiple real-time identification in one projection process and the agreed criteria.
Description
Technical Field
The invention relates to a cinema monitoring technology, in particular to a method for counting cinema audience numbers by integrating video monitoring and radio frequency identification.
Background
In recent years, the movie industry in China rapidly develops, but the phenomenon of 'steal conceal' ticket houses in cinema still exists objectively. In the movie market, in order to prevent the occurrence of the phenomenon of ticket house faking, the number of audiences in each showing process is generally required to be counted, so that accurate statistics is made for ticket houses of movie works, and a manufacturer, a cinema owner, the audiences and the like can conveniently make correct judgment on the quality and the value of the movie works.
The conventional ticket house supervision means is mainly to manually count the sales condition of the tickets, and the method has low efficiency and is not easy to ensure the authenticity of the ticket house. Radio frequency identification and video surveillance are often used to count the number of viewers.
Video monitoring is a device for intelligently identifying audiences by utilizing deep learning, and comprises an up-and-down swinging bracket and a camera part, wherein the camera is installed above a cinema seat, and the seat of the whole cinema can be shot by adjusting the camera through the up-and-down bracket. During the cinema showing process, the computer recognizes the photographed image, and the audience detection is completed. When the shot image is too bright and too dark, the result of the video monitoring detection may be inaccurate.
Radio frequency identification is a wireless communication technology, and comprises a reader-writer, a reader-writer antenna, a tag and the like, wherein the reader-writer radiates electromagnetic waves to the tag through the reader-writer antenna, provides energy for the tag, and sends a command to the tag. After the tag enters the working range of the reader-writer antenna, energy is obtained to start working, the tag returns the identification and other information of the tag to the reader-writer antenna, and the reader-writer completes identification of the tag. In the method for counting audience number of theatres by using radio frequency identification 201911290776.0 disclosed in the publication, when liquid or metal interferents are placed on a seat, the result of radio frequency identification is inaccurate, and at the moment, the video monitoring technology is fused, so that the defect of radio frequency identification can be overcome.
Therefore, if only video monitoring is considered, the condition that the shot image is blurred or excessively bright and excessively dark may occur; if only the radio frequency identification technology is considered, interference may be present. The method combines video monitoring and radio frequency identification, complements the gap, realizes double accurate statistics, and can meet the requirements of movie market supervision. The present invention has been made to meet such a real demand.
Disclosure of Invention
The invention aims to provide a method for counting cinema audience numbers by integrating video monitoring and radio frequency identification, which is used for solving the problems in the prior art.
The invention discloses a method for counting cinema audience numbers by integrating video monitoring and radio frequency identification, which comprises the following steps: (1) Installing a radio frequency identification reader-writer, an antenna, a camera and a computer; (2) The radio frequency identification tag which uniquely identifies each seat is arranged in the seat surface of the cinema seat, so that the identification of all seats of the cinema is realized, under the condition of no person sitting, a computer controls a camera to shoot an image, each seat in the image is calibrated, then the image of each seat is intercepted according to the coordinates of each seat in the image, and the reference correction is carried out; (3) The method comprises the steps that seat images with audiences or not are input into a deep neural network for training, a deep neural network model for judging whether the seats are occupied or not is obtained, under the condition that no seat is occupied, a reader-writer identifies all the tags of seats in a cinema, the signal intensity range of radio frequency identification tags of all the seats is counted, and a signal intensity model is built; (4) At the appointed moment in the projection process, the computer controls the camera to shoot images, inputs a deep neural network model, accurately identifies whether audience seats exist or not, counts the number of audiences, and calculates the number of audiences according to the number of tags obtained by reference identification and the number of tags obtained by multiple real-time identification in one projection process and the agreed criterion.
According to one embodiment of the method, a camera is installed in a cinema, a support is fixed to a ceiling, the camera is fixed to the support, the support knob is adjusted by comparing images on video monitoring, all seats can be shot by adjusting angles, images shot by the camera can be successfully transmitted to a computer, and interconnection and intercommunication between the camera and the computer are achieved.
According to an embodiment of the method of the present invention, one reader is connected to a plurality of antennas of one theater through a cable, and one computer is connected to a plurality of readers through a network.
According to an embodiment of the method of the invention, the radio frequency identification tag is an ultra-high frequency tag of 800/900MHz band.
An embodiment of the method according to the invention, wherein the convolutional neural network structure: the input image size of the first input layer is 224 x 3, the second convolution layer uses a convolution kernel with a sliding step length of 2 of 7*7 channels, the third maximum pooling layer uses a convolution kernel with a sliding step length of 2 of 3*3, the fourth layer is a convolution residual error module, and the fifth layer is an average pooling layer and a soft maximum output layer.
An embodiment of the method according to the invention is defined by the average pooling layer and a 1000-dimensional full link soft maximum output layer for distinguishing 1000 different categories.
An embodiment of the method according to the invention, wherein the convolution residual module comprises four types: the first comprises three convolutions of conv2_1[1, 64], conv2_2[3, 64], conv2_3[1, 256] with outputs of 56 x 256, the second comprises three convolutions of conv3_1[1, 128], conv3_2[3, 128] and conv3_3[1 x1, 512] with outputs of 28 x 512, the third comprises three convolutions of conv4_1[1, 256], conv4_2[3 x3, 256] and conv4_3[1, 1024] with outputs of 14 x 1024, the fourth comprises three convolutions of conv5_1[1, 256], conv5_2[3 x3, 512] and conv5_3[1, 8] with outputs of 2048 x 7; the four convolution residual modules have convolution neural network structures at the same time or have partial convolution neural network structures.
According to an embodiment of the method, a computer controls a camera to shoot cinema seat images and inputs a neural network to identify whether a person exists in a seat; and controlling the reader-writer to identify the tag in one cinema, and according to the signal intensity models of the tags of the seats, if the signal intensity of the tag is in the range of the signal intensity model, considering that the tag is not shielded or abnormal, and the seats are idle and unmanned.
According to an embodiment of the method of the present invention, wherein for video monitoring the average number of viewers is calculated using the number of viewers = number of viewers/number of images identified by a plurality of images, and the number of theatres is calculated using the number of viewers = number of reference tags-number of minimum free seat tags, or the number of viewers = number of reference tags-number of average free seat tags.
An embodiment of the method according to the invention, further comprises: determining the number of spectators includes: if the computer controls the camera to shoot the cinema seat image, and inputs the cinema seat image into the neural network to identify the seat unmanned, and the seat unmanned is judged by the mode that the signal intensity of the tag is in the range of the signal intensity model, the seat unmanned is judged; otherwise, it is determined that the seat is occupied.
Aiming at the conditions of blurring or excessive brightness, excessive darkness and the like of a video monitoring image, the invention uses the radio frequency identification to supplement the defects; aiming at the situations that the radio frequency identification is affected by the interference objects placed on the seat, the video monitoring is used for identifying the deficiency. The fusion of the two statistical results requires the elimination of the respective defects before the presence or absence of the audience can be determined. The real-time double accurate statistics of the number of audiences in the cinema is realized, and the authenticity of the ticket houses and the benefits of all parties are guaranteed. The invention plays an important role in the field of performance market supervision.
Drawings
FIG. 1 is a flow chart of a method of integrating video surveillance and radio frequency identification to count the number of theatre audience members of the present invention;
Fig. 2 is a schematic diagram of the method of the present invention for integrating video surveillance and rfid statistics of the number of theatre audience.
Detailed Description
For the purposes of clarity, content, and advantages of the present invention, a detailed description of the embodiments of the present invention will be described in detail below with reference to the drawings and examples.
The invention discloses a method for counting cinema audience numbers by integrating video monitoring and radio frequency identification. The method comprises the following steps: (1) And a step of installing the radio frequency identification reader-writer, the antenna, the camera and the computer. The reader-writer antenna is arranged above the cinema seat, the camera capable of identifying the whole cinema seat is arranged on the cinema ceiling, the reader-writer and the computer are arranged at other positions of the cinema, and the reader-writer antenna, the reader-writer, the camera and the computer are connected, so that the computer can control the camera and the reader-writer to identify audience and labels on the cinema seat. (2) a radio frequency identification tag mounting step. And a radio frequency identification tag which uniquely identifies each seat is arranged in the seating surface of the cinema seat, so that the identification of all seats of the cinema is realized. Under the condition of normal seating, the computer controls the camera to shoot an image, each seat in the image is calibrated in a manual mode, then the image of each seat is intercepted according to the coordinates of each seat in the image, and reference correction is carried out. (3) a reference recognition step. And inputting the normalized seat images with or without audience into a deep neural network for training to obtain a deep neural network model for judging whether the seat is occupied or not. Under the condition of normal seating, the reader-writer identifies the tags of all seats in one cinema, and establishes a signal intensity model of each seat tag. (4) a real-time statistics step. The time is specified in the showing process: the computer controls the camera to shoot images, inputs a deep neural network model, realizes accurate identification of whether audience seats exist or not, and counts the number of audiences. The computer calculates the audience number according to the standard identification tag number and the tag number obtained by multiple real-time identification in one showing process and the agreed criterion. (5) a double statistical fusion step. If the numbers of the audience identified by the two technologies are inconsistent, the actual number of the audience is judged according to a certain criterion.
The double statistical fusion step specifically comprises: the number of viewers is determined according to certain criteria. If both technologies judge that no person is on a certain seat, judging that the seat is unmanned; if both techniques determine that a person is present in a seat, then determining that the seat is present; if the radio frequency identification technology judges that a certain seat is unmanned, the video monitoring technology judges that the seat is unmanned, and the video misidentification is caused by too bright and too dark light; if the radio frequency identification technology judges that a person is on a certain seat, the video monitoring technology judges that the person is not on the seat, and the radio frequency identification technology judges that an interfering object (such as liquid or metal) is placed on the seat, and judges that the person is not on the seat.
With reference to fig. 1 and 2, the implementation flow of the present invention is as follows;
(1) And a step of installing the radio frequency identification reader-writer, the antenna, the camera and the computer. The reader-writer antenna is arranged above the cinema seat, the camera capable of identifying the whole cinema seat is arranged on the cinema ceiling, the reader-writer and the computer are arranged at other positions of the cinema, and the reader-writer antenna, the reader-writer, the camera and the computer are connected, so that the computer can control the camera and the reader-writer to identify audience and labels on the cinema seat.
In practice, a cinema mounts a camera, secures the stand to the ceiling, and secures the camera to the stand. By comparing images on video monitoring, the bracket knob is adjusted, and all seats can be shot by adjusting angles. Ensure that the image shot by the camera can be successfully transmitted to the computer, realize the interconnection and intercommunication between the camera and the computer,
And a plurality of reader-writer antennas are arranged, the action ranges among the antennas can be partially overlapped, and the action ranges of the plurality of reader-writer antennas need to cover the whole cinema, so that the tags in the cinema are prevented from being read out. One reader-writer can be connected with a plurality of antennas of a cinema through cables, and one computer can be connected with a plurality of reader-writers through a network.
(2) And a radio frequency identification tag mounting step. And a radio frequency identification tag which uniquely identifies each seat is arranged in the seating surface of the cinema seat, so that the identification of all seats of the cinema is realized.
In the specific implementation, the radio frequency identification tag adopts an ultra-high frequency tag with the frequency band of 800/900MHz, and the identification distance is long without a battery. The label adopts the reinforcement design, avoids being crushed. The tag is mounted on the seating surface of the seat such that when an object is on the seat, the object can shield the radio frequency identification tag and the reader cannot identify the tag. The identification in the tag can be realized by adopting a mode of combining cinema numbers and seat numbers.
Under the unmanned condition, the computer controls the camera to shoot an image, each seat in the image is calibrated in a manual mode, then the image of each seat is intercepted according to the coordinates of each seat in the image, and reference correction is carried out.
(3) And a reference identification step. And inputting the normalized seat images with or without audience into a deep neural network for training to obtain a deep neural network model for judging whether the seat is occupied or not. Under the condition of no person sitting, the reader-writer identifies the tags of all seats in one cinema, counts the signal intensity range of the radio frequency identification tags of all seats, and establishes a signal intensity model.
In particular, the convolutional neural network structure is that a first input layer inputs an image with 224 x3, a second convolutional layer uses 7*7 convolutional kernels with 64 channels and a sliding step length of 2, a third layer maximum pooling layer uses a convolutional kernel with 3*3 and a sliding step length of 2, a fourth layer is a convolutional residual module, a first one comprises three convolutional layers of conv2_1[1 x 1,64], conv2_2[3 x3, 64], conv2_3[1 x 1,256], the output of which is 56 x 256, a second one comprises three convolutional layers of conv3_1[1 x 1,128], conv3_2[3 x3, 128], conv3_3[1 x 1,512], the output is 28 x 512, the third one contains conv4_1[1 x 1,256], conv4_2[3 x3, 256], conv4_3[1 x 1,1024] three convolution layers, the output is 14 x 1024, the fourth one contains conv5_1[1 x 1,256], conv5_2[3 x3, 512], conv5_3[1 x 1,2048] three convolution layers, the output is 7 x 2048, the fifth one is an average pooling layer and a soft maximum output layer, the average pooling layer and a 1000-dimensional full link soft maximum output layer, for distinguishing 1000 different categories.
The camera can identify all seats, and the computer records the label identifications of all seats of each cinema, so that the labels of all seats are identified.
(4) And (5) carrying out real-time statistics. The time is specified in the showing process: the computer controls the camera to shoot images, inputs a deep neural network model, realizes accurate identification of whether audience seats exist or not, and counts the number of audiences. The computer calculates the audience number according to the standard identification tag number and the tag number obtained by multiple real-time identification in one showing process and the agreed criterion.
In the implementation, the camera is controlled by the computer to shoot the cinema seat image at a plurality of moments in the projection process, such as 5 minutes after the start, 5 minutes in the middle and 5 minutes before the end, and the cinema seat image is input into the neural network to identify whether a person exists in the seat. And meanwhile, the reader-writer is controlled to identify the tag in one cinema, and if the signal intensity of the tag is in the range of the signal intensity model according to the signal intensity model of each seat tag, the tag is considered to be not shielded and abnormal, and the seats are idle and unmanned.
For video monitoring, the average number of viewers can be calculated by adopting the mode of number of viewers=number of viewers identified by a plurality of images/number of images. The theatre audience count may be calculated using the number of audience = reference number of tags-minimum number of idle seat tags, or the number of audience = reference number of tags-average number of idle seat tags, or the like.
(5) And (5) double statistical fusion step. The number of viewers is determined according to certain criteria. If both technologies judge that no person is on a certain seat, judging that the seat is unmanned; if both techniques determine that a person is present in a seat, then determining that the seat is present; if the radio frequency identification technology judges that a certain seat is unmanned, the video monitoring technology judges that the seat is unmanned, and the video misidentification is caused by too bright and too dark light; if the radio frequency identification technology judges that a person is on a certain seat, the video monitoring technology judges that the person is not on the seat, and the radio frequency identification technology judges that an interfering object (such as liquid or metal) is placed on the seat, and judges that the person is not on the seat.
The invention identifies and marks the seats in the cinema, identifies whether audience exists on the seats, and realizes double accurate statistics on the number of the audience, thereby protecting the fairness of the ticket houses and being beneficial to accurately judging the quality and the value of film and television works by producers, cinema owners, audience and the like. Therefore, the invention will play an important role in the field of movie market supervision.
The foregoing is merely a preferred embodiment of the present invention, and it should be noted that modifications and variations could be made by those skilled in the art without departing from the technical principles of the present invention, and such modifications and variations should also be regarded as being within the scope of the invention.
Claims (5)
1. A method for integrating video surveillance and radio frequency identification to count the number of theatre audience, comprising:
(1) Installing a radio frequency identification reader-writer, an antenna, a camera and a computer;
(2) The radio frequency identification tag which uniquely identifies each seat is arranged in the seat surface of the cinema seat, so that the identification of all seats of the cinema is realized, under the condition of no person sitting, a computer controls a camera to shoot an image, each seat in the image is calibrated, then the image of each seat is intercepted according to the coordinates of each seat in the image, and the reference correction is carried out;
(3) The method comprises the steps that seat images with audiences or not are input into a deep neural network for training, a deep neural network model for judging whether the seats are occupied or not is obtained, under the condition that no seat is occupied, a reader-writer identifies all the tags of seats in a cinema, the signal intensity range of radio frequency identification tags of all the seats is counted, and a signal intensity model is built;
(4) The method comprises the steps that at a designated moment in the projection process, a computer controls a camera to shoot images, a deep neural network model is input, whether audience seats exist or not is accurately identified, the number of audiences is counted, and the computer calculates the number of audiences according to the number of tags obtained through reference identification and the number of tags obtained through multiple real-time identification in one projection process and a stipulated criterion;
Wherein,
Convolutional neural network structure: the input image of the first input layer has 224 x 3, the second convolution layer uses a convolution kernel with the sliding step length of 7*7 channels of 64 channels of 2, the third maximum pooling layer uses a convolution kernel with the sliding step length of 3*3 of 2, the fourth layer is a convolution residual error module, and the fifth layer is an average pooling layer and a soft maximum output layer;
The convolution residual modules include four types:
The first comprises three convolutions of conv2_1[1, 64], conv2_2[3, 64], conv2_3[1, 256] with outputs of 56 x 256, the second comprises three convolutions of conv3_1[1, 128], conv3_2[3, 128] and conv3_3[ 1x1, 512] with outputs of 28 x 512, the third comprises three convolutions of conv4_1[1, 256], conv4_2[ 3x 3,256] and conv4_3[1, 1024] with outputs of 14 x 1024, the fourth comprises three convolutions of conv5_1[1, 256], conv5_2[ 3x 3,512] and conv5_3[1, 8] with outputs of 2048 x 7; the four convolution residual modules have convolution neural network structures at the same time or have partial convolution neural network structures;
The computer controls the camera to shoot the cinema seat image and inputs the cinema seat image into the neural network to identify whether a person exists in the seat; the reader-writer is controlled to identify the tag in one cinema, and according to the signal intensity model of each seat tag, if the signal intensity of the tag is in the range of the signal intensity model, the tag is considered to be not shielded and abnormal, and the seats are idle and unmanned;
When the number of people determined in the two ways is inconsistent, determining the number of audience comprises:
If both technologies judge that no person is on a certain seat, judging that the seat is unmanned; if both techniques determine that a person is present in a seat, then determining that the seat is present; if the radio frequency identification technology judges that a certain seat is unmanned, the video monitoring technology judges that the seat is unmanned, and the video misidentification is caused by too bright and too dark light; if the radio frequency identification technology judges that a person is on a certain seat, the video monitoring technology judges that the person is not on the seat, and if the fact that an interfering object is placed on the seat is indicated, the person is judged to be not on the seat;
For video monitoring, calculating an average audience number by adopting the audience number = the audience number/the image number identified by a plurality of images;
the theatre audience number is calculated for radio frequency identification using the audience number = reference tag number-minimum number of idle seat tags, or the audience number = reference tag number-average number of idle seat tags.
2. The method of claim 1 wherein a cinema mounts a camera, secures the bracket to the ceiling, secures the camera to the bracket, adjusts the bracket knob by comparing images on video monitors, adjusts the angle to capture all seats, ensures that images captured by the camera can be successfully transferred to the computer, and enables interconnection of the camera and the computer.
3. The method of claim 2, wherein a reader is connected to a plurality of antennas of a theater via a cable, and a computer is connected to a plurality of readers via a network.
4. The method of claim 1, wherein the radio frequency identification tag is an ultra-high frequency tag of 800/900MHz band.
5. The method of claim 1, wherein an average pooling layer and a 1000-dimensional full link soft maximum output layer are used to distinguish 1000 different categories.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010991479.5A CN112149768B (en) | 2020-09-17 | 2020-09-17 | Method for counting cinema audience number by integrating video monitoring and radio frequency identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010991479.5A CN112149768B (en) | 2020-09-17 | 2020-09-17 | Method for counting cinema audience number by integrating video monitoring and radio frequency identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112149768A CN112149768A (en) | 2020-12-29 |
| CN112149768B true CN112149768B (en) | 2024-05-14 |
Family
ID=73893300
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010991479.5A Active CN112149768B (en) | 2020-09-17 | 2020-09-17 | Method for counting cinema audience number by integrating video monitoring and radio frequency identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112149768B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113014756A (en) * | 2021-02-02 | 2021-06-22 | 刘毅 | Hidden camera shooting mechanism driving system |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20070004294A (en) * | 2005-07-04 | 2007-01-09 | 정제창 | System and method for counting audience number |
| CN101188743A (en) * | 2007-09-17 | 2008-05-28 | 深圳先进技术研究院 | An intelligent digital system based on video and its processing method |
| JP2011004050A (en) * | 2009-06-17 | 2011-01-06 | Panasonic Corp | Monitoring camera system and recorder |
| CN103345653A (en) * | 2013-06-17 | 2013-10-09 | 复旦大学 | Attendance statistical method based on multi-camera integration |
| CN103942850A (en) * | 2014-04-24 | 2014-07-23 | 中国人民武装警察部队浙江省总队医院 | Medical staff on-duty monitoring method based on video analysis and RFID (radio frequency identification) technology |
| CN104809909A (en) * | 2015-03-05 | 2015-07-29 | 桑田智能工程技术(上海)有限公司 | Cell empty carport identification system based on RFID signal intensity and assistant camera video |
| CN104899556A (en) * | 2015-05-14 | 2015-09-09 | 中国农业大学 | System for counting persons in classroom based on image recognition |
| CN105447880A (en) * | 2015-12-15 | 2016-03-30 | 北京文安科技发展有限公司 | Box-office attendance verification method, device and system |
| CN107122698A (en) * | 2016-07-19 | 2017-09-01 | 安徽大学 | A kind of real-time attendance statistical method of cinema based on convolutional neural networks |
| CN107341538A (en) * | 2017-06-30 | 2017-11-10 | 上海上大海润信息系统有限公司 | A kind of statistical magnitude method of view-based access control model |
| CN107610477A (en) * | 2017-09-22 | 2018-01-19 | 深圳鼎识科技有限公司 | A kind of intelligent transportation radio frequency and video information method for amalgamation processing |
| JP2019200533A (en) * | 2018-05-15 | 2019-11-21 | パナソニックＩｐマネジメント株式会社 | Counting device, accounting system, learning device, and control method |
| CN210274297U (en) * | 2019-04-11 | 2020-04-07 | 金色大田科技有限公司 | Livestock breeding population quantity monitoring system |
| CN111126090A (en) * | 2019-12-16 | 2020-05-08 | 北京计算机技术及应用研究所 | Method for counting number of theatre audiences by utilizing radio frequency identification |
-
2020
- 2020-09-17 CN CN202010991479.5A patent/CN112149768B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20070004294A (en) * | 2005-07-04 | 2007-01-09 | 정제창 | System and method for counting audience number |
| CN101188743A (en) * | 2007-09-17 | 2008-05-28 | 深圳先进技术研究院 | An intelligent digital system based on video and its processing method |
| JP2011004050A (en) * | 2009-06-17 | 2011-01-06 | Panasonic Corp | Monitoring camera system and recorder |
| CN103345653A (en) * | 2013-06-17 | 2013-10-09 | 复旦大学 | Attendance statistical method based on multi-camera integration |
| CN103942850A (en) * | 2014-04-24 | 2014-07-23 | 中国人民武装警察部队浙江省总队医院 | Medical staff on-duty monitoring method based on video analysis and RFID (radio frequency identification) technology |
| CN104809909A (en) * | 2015-03-05 | 2015-07-29 | 桑田智能工程技术(上海)有限公司 | Cell empty carport identification system based on RFID signal intensity and assistant camera video |
| CN104899556A (en) * | 2015-05-14 | 2015-09-09 | 中国农业大学 | System for counting persons in classroom based on image recognition |
| CN105447880A (en) * | 2015-12-15 | 2016-03-30 | 北京文安科技发展有限公司 | Box-office attendance verification method, device and system |
| CN107122698A (en) * | 2016-07-19 | 2017-09-01 | 安徽大学 | A kind of real-time attendance statistical method of cinema based on convolutional neural networks |
| CN107341538A (en) * | 2017-06-30 | 2017-11-10 | 上海上大海润信息系统有限公司 | A kind of statistical magnitude method of view-based access control model |
| CN107610477A (en) * | 2017-09-22 | 2018-01-19 | 深圳鼎识科技有限公司 | A kind of intelligent transportation radio frequency and video information method for amalgamation processing |
| JP2019200533A (en) * | 2018-05-15 | 2019-11-21 | パナソニックＩｐマネジメント株式会社 | Counting device, accounting system, learning device, and control method |
| CN210274297U (en) * | 2019-04-11 | 2020-04-07 | 金色大田科技有限公司 | Livestock breeding population quantity monitoring system |
| CN111126090A (en) * | 2019-12-16 | 2020-05-08 | 北京计算机技术及应用研究所 | Method for counting number of theatre audiences by utilizing radio frequency identification |
Non-Patent Citations (2)
| Title |
|---|
| 基于射频识别技术的地铁车厢人数检测方法研究;张超宇;;科技风(第11期);第115+117页 * |
| 张超宇 ; .基于射频识别技术的地铁车厢人数检测方法研究.科技风.2016,(第11期),第115+117页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112149768A (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108419023B (en) | Method for generating high dynamic range image and related equipment | |
| Li et al. | Real-time screen-camera communication behind any scene | |
| CN109714519B (en) | Method and system for automatically adjusting image frame | |
| CN110769246B (en) | Method and device for detecting faults of monitoring equipment | |
| CN101344988B (en) | Image acquisition and processing equipment and method, vehicle monitoring and recording system | |
| CN111770285B (en) | Exposure brightness control method and device, electronic equipment and storage medium | |
| CN110890078B (en) | Method for automatically adjusting brightness of display screen based on pupil size | |
| CN104168478B (en) | Based on the video image color cast detection method of Lab space and relevance function | |
| CN106210471A (en) | A kind of outdoor face recognition method and system | |
| CN103150584B (en) | Communication resource moving processing method and system | |
| CN105407276A (en) | Photographing method and equipment | |
| CA2538301A1 (en) | Computerized method and apparatus for determining field-of-view relationships among multiple image sensors | |
| CN112149768B (en) | Method for counting cinema audience number by integrating video monitoring and radio frequency identification | |
| CN110087049A (en) | Automatic focusing system, method and projector | |
| CN106469292A (en) | License plate recognition control system and its identification control method | |
| CN108805184A (en) | A kind of fixed space, image-recognizing method and system on vehicle | |
| CN115147362A (en) | Detection method, detection device and detection system for display panel | |
| CN201247528Y (en) | Apparatus for obtaining and processing image | |
| CN105225254B (en) | A kind of exposure method and system of automatic tracing localized target | |
| CN108391044A (en) | A kind of video camera and its mode switch control method, device | |
| CN112183158B (en) | Cereal type identification method of cereal cooking equipment and cereal cooking equipment | |
| CN107995476A (en) | A kind of image processing method and device | |
| CN111160299A (en) | Living body identification method and device | |
| CN113301315B (en) | Projection system based on infrared touch screen frame | |
| CN108647615A (en) | A kind of cinema steals a ride the implementation method of early warning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |