WO2015084148A1 - A system and method for extracting objects from interlaced images - Google Patents
A system and method for extracting objects from interlaced images Download PDFInfo
- Publication number
- WO2015084148A1 WO2015084148A1 PCT/MY2014/000162 MY2014000162W WO2015084148A1 WO 2015084148 A1 WO2015084148 A1 WO 2015084148A1 MY 2014000162 W MY2014000162 W MY 2014000162W WO 2015084148 A1 WO2015084148 A1 WO 2015084148A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- images
- time
- path
- travel time
- optical sensor
- 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.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Definitions
- This invention is related to the field of processing of multiplexed images and more particularly a system and method for extracting objects from interlaced images.
- US 6240217 B1 describes a method for processing digital image information including pixel intensity information that selectively modifies pixel intensity information in order to improve legibility or visibility of parts of a digital image.
- pixel intensity information that selectively modifies pixel intensity information in order to improve legibility or visibility of parts of a digital image.
- Both of these prior arts emphasizes on the analysis of images based on the digital information obtained from the captured image but the prior arts did not address the issue of identifying vehicles with similar physical features from multiplexed snapshots or images. Also, the prior arts did not address the issue of analyzing motor vehicles that may be present in interlaced images.
- the present invention relates to a system and method for extracting objects from interlaced images.
- the system as disclosed in one of the embodiment of the present invention comprises a video image acquisition module that captures images of a moving object along a path of interest using optical sensors a moving object detection module that differentiates movement of the object from a static background of the images captured a visualization display that shows the captured images of the object, which the system is characterized by a calibration data profiling module that represents connections of the optical sensors with each other and the paths to be taken by the object and derives the mean of the travel time of the object based on recorded event time of the object detected, and a delayed real-time detection module that calculates the time taken by the object to travel along the said path based on real-time delayed detection, and compares the said time with the mean of the travel time of the object based on the recorded event time of the object detected.
- the method as disclosed in a further embodiment of the present invention comprises the steps of generating a path topology map to represent connections of optical sensors with each other, computing a calibrated event profile module to derive the mean of travel time of the object based on time records of the object detected, acquiring video images of the object moving along the paths, detecting moving object, deriving a travel time of the object along each of the path based on real-time delayed detection, comparing the travel time of the object along each of the paths based on real-time delayed detection with the travel time of the object based on means of the calibrated event profile, extracting and grouping the video images of the objects into categories for display, and displaying a timeline of the images of the object based on the categories.
- Figure 1 illustrates the flow diagram of the method for extracting objects from interlaced images.
- Figure 2 illustrates the flow diagram of the step of generating a path topology map.
- Figure 3 illustrates the generation of a path topology map and a look-up table for the path topology map.
- Figure 4 illustrates the flow diagram of the step of computing a calibrated event profile module.
- Figure 5 illustrates the calibrated event profile for each path in the path topology map.
- Figure 6 illustrates the calibrated travel time for each respective path in the path topology map.
- Figure 7 illustrates the mean of calibrated travel time for each path.
- Figure 8 illustrates the flow diagram of the step of acquiring video images of the object moving along the paths.
- Figure 9 illustrates the extraction of video streams into frames.
- Figure 10 illustrates the flow diagram of the step of detecting the presence of moving objects.
- Figure 11 illustrates the scanning of the frames to differentiate a moving object from background image.
- Figure 12 illustrates the flow diagram of the step of deriving the travel time of the object along each of the path based on real-time delayed detection.
- Figure 13 illustrates the computation of the travel time of the moving object.
- Figure 14 illustrates the flow diagram of the step of comparing the travel time of the object.
- Figure 15 illustrates the computation of the value of the difference in percentage (%) and the comparison of the value with the threshold having a range of ⁇ 2.5%.
- Figure 16 illustrates the flow diagram of the step of extracting and grouping the images of the objects into categories for display.
- Figure 17 illustrates the storing of images in display buffers.
- Figure 18 illustrates the flow diagram of the step of displaying the timeline of the images of the object based on the categories.
- Figure 19 illustrates the display of the images in a timeline.
- One of the embodiment of the present invention illustrates a system for extracting objects from interlaced images comprising of a video image acquisition module that captures images of a moving object along a path of interest using optical sensors a moving object detection module that differentiates movement of the object from a static background of the images captured a visualization display that shows the captured images of the object, which is characterized by a calibration data profiling module that represents connections of the optical sensors with each other and the paths to be taken by the object and derives the mean of the travel time of the object based on recorded event time of the object detected, and a delayed real-time detection module that calculates the time taken by the object to travel along the said path based on real-time delayed detection, and compares the said time with the mean of the travel time of the object based on the recorded event time of the object detected.
- the calibration data profiling module generates a path topology map to represent the connections of the optical sensors with each other and the paths to be taken by the object. It is disclosed in the present embodiment of the present invention that the mean of travel time is changeable to any time period of the day and week.
- a further embodiment of the present invention illustrates a method for extracting objects from interlaced images comprising the steps of generating a path topology map to represent connections of optical sensors with each other, computing a calibrated event profile module to derive the mean of travel time of the object based on time records of the object detected, acquiring video images of the object moving along the paths, detecting moving object, deriving a travel time of the object along each of the path based on real-time delayed detection, comparing the travel time of the object along each of the paths based on real- time delayed detection with the travel time of the object based on means of the calibrated event profile, extracting and grouping the video images of the objects into categories for display, and displaying a timeline of the images of the object based on the categories.
- the categories mentioned herein are determined based on similarities of the images and the images are grouped into their respective categories based on the similarities that occur in the images.
- the steps described herein are illustrated in Figure 1.
- the step of generating the path topology map further comprises the steps of retrieving index numbers of the optical sensors and respective location of the optical sensors within the area of interest from a first database, setting a path to any two of the optical sensors that are connected to each other, generating a path topology map to represent the said paths, transforming the path topology map into a look-up table, and storing the path topology map and the look-up table.
- the optical sensor is identified as CamlD. If one of the optical sensor is named CamA and the other optical sensor is named CamB, the path is set with either CamA_CamB and CamB_CamA.
- the steps described herein are illustrated in Figure 2.
- the path topology map and look-up table are illustrated in Figure 3.
- the second step of the method which is the computation of the calibrated event profile module further comprising the steps of retrieving a period of time of the object's recorded timestamp for each optical sensor from a second database, searching for paths that are connected to the optical sensors from the path topology map, computing a travel time of the object for each path in a calibrated event profile module, calculating a mean of calibrated travel times for each path, and storing the mean calibrated travel time into a third database.
- all information related to the particular optical sensor is also retrieved. From the path topology map, the paths that are connected to the particular optical sensor is determined when the path of CamA_CamB shows TRUE or the path of CamB_CamA shows TRUE.
- Figure 5 illustrates the calibrated event profile for each path in the path topology map and Figure 6 illustrates the calibrated travel time for each respective path in the path topology map.
- the calibration method is carried with constraints such that the object's speed is within a speed limit, the object is not allowed to stop midway of the path, the rear part of the object is not allowed to overtake the front part of the object along the path, and the path is a single lane path.
- the computation switches to a subsequent optical sensor having the object's timestamp. If the subsequent optical sensor does not have any timestamp, the computation proceeds with the calculation of the mean of calibrated travel times for each path.
- Figure 7 illustrates the mean of calibrated travel time for each path.
- the computation Before storing the mean of calibrated travel times for each path, the computation also determines if a subsequent path exists. If the subsequent path exists, the steps of retrieving timestamps and information of the optical sensors and the following steps are repeated until no subsequent path exists.
- the steps described herein are illustrated in Figure 4. Subsequently, video images are acquired and the step of acquiring the video images further comprises the steps of retrieving index numbers of the optical sensors and locations within the area of interest from a first database, determining video stream availability from the optical sensors, extracting the video streams into frames, and storing the frames into video buffers.
- Figure 8 illustrates the flow diagram of the steps described herein and Figure 9 illustrates the extraction of video streams into frames.
- the step of detecting the presence of moving objects further comprises the steps of retrieving frames from video buffers according to the index numbers of the optical sensors, scanning the frames to differentiate a moving object from background image, extracting frames having the moving object together with a recorded timestamp and index number of the optical sensor, and updating the recorded timestamp and index number in a camera event timestamp buffer.
- the current optical sensor's index has been detected and recorded previously, the previous detected and recorded time with the last value is overwritten.
- the step of deriving the travel time of the object along each of the path based on real-time delayed detection further comprises the steps of retrieving an index number and recorded timestamp of a first optical sensor that is connected to one end of a path from a camera event timestamp buffer, retrieving an index number and recorded timestamp a second optical sensor that is connected to the other end of the said path from the camera event timestamp buffer, computing a difference between the recorded timestamps from the first optical sensor and the second optical sensor to obtain a travel time of the moving object, and storing the travel time of the moving object into a fourth database.
- Figure 12 illustrates the steps described herein and Figure 13 illustrates the computation of the travel time of the moving object.
- the travel times of the object are compared by retrieving information of the optical sensor where the moving object is detected, retrieving mean calibrated travel time from a third database, retrieving travel time of the moving object based on real-time delayed detection from a fourth database, computing differences between the travel time obtained from real-time delayed detection and the mean calibrated travel time to determine whether the object is the same object or a different object, determining whether the computed difference has a value that falls within a threshold, and determining validity of the value of the difference.
- the value of the difference computed is in percentage (%) and the value is compared with the threshold having a range of ⁇ 2.5%. The validity of the said value is determined based on whether the value falls within this range.
- Figure 14 illustrates the steps described herein and Figure 15 illustrates the computation of the value of the difference in percentage (%) and the comparison of the value with the threshold having a range of ⁇ 2.5%.
- the step of extracting and grouping video images of the objects further comprises the steps of determining whether the same object is detected, obtaining information of the optical sensor, obtaining images of the detected object based on the real-time delayed detection using the recorded timestamp, determining whether category of the object exists, and storing the information of the optical sensor and the images of the object into a first display buffer when the category of the object exists, wherein the information of the optical sensor and the images of the object is stored into a second display buffer when the category does not exists.
- Figure 16 illustrates the steps described herein and Figure 17 illustrates the storing of images in display buffers.
- the step of displaying the timeline of the video images of the object based on the categories further comprises the steps of determining whether there is any update from display buffers, obtaining sequential images of the event detection and counter checking the sequential images with the information of the recorded time for the arrangement of the images in the order of time, sending the images to a display monitor, determining whether a display buffer is generated, and repeating the step of obtaining and counter checking of the sequential images and sending the images to the display monitor.
- the arrangement of the images is in the order of time forming a timeline and the timeline is arranged in sequence and categories of the images.
- Figure 18 illustrates the steps described herein and Figure 19 illustrates the display of the images in a timeline.
- the object as described is a vehicle, which the system and method may be used for detecting a vehicle or similar vehicles in interlaced images.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The present invention relates to a system and method for extracting objects from interlaced images, wherein the system and method generates a path topology map and computes a calibrated event profile module to identify similar objects moving sequentially in the same direction along the same path and similar objects moving sequentially in different directions along the same path in interlaced images. Most illustrative drawing: Figure
Description
A SYSTEM AND METHOD FOR EXTRACTING OBJECTS FROM
INTERLACED IMAGES
TECHNICAL FIELD OF THE INVENTION
This invention is related to the field of processing of multiplexed images and more particularly a system and method for extracting objects from interlaced images.
BACKGROUND OF THE INVENTION
A huge number of apparatuses, methods and systems have been used to capture and identify motor vehicles in flowing traffic. These apparatuses, methods and systems are crucial tools for traffic management and enforcement teams to monitor traffic flow and to enforce the law of traffic to infringers. Most of these apparatuses, methods and systems capture images of the motor vehicles using cameras and sensors, and identify the motor vehicles based on their external or physical features, for instance, license plates of the motor vehicles.
However, a number of problems have been identified and researchers have been actively involved in solving the problems and improving the apparatuses, methods and systems in order to manage traffic flow in a better manner.
Due to cognitive overload and short attention span of users, conventional apparatuses, methods and systems are unable to identify vehicles with similar physical features from multiplexed snapshots or images. Conventionally, physical features such as the motor vehicles' dominant colours or license plate contents are used as matching reference to identify motor vehicles or similar motor vehicles. However, the problem of identifying and extracting wrong motor vehicle may occur due to factors such as poor lighting condition, which deters the motor vehicles' colour extraction process, and noise and glare in the distorted license plate images of the motor vehicles captured.
US 20130050492 A1 describes a method and apparatus for identifying motor vehicles for monitoring traffic. The prior art analyses and determines a motor vehicle from the size ratios of a license-plate contour in a distorted image. US 6240217 B1 describes a method for processing digital image information including pixel intensity information that selectively modifies pixel intensity information in order to improve legibility or visibility of parts of a digital image. Both of these prior arts emphasizes on the analysis of images based on the digital information obtained from the captured image but the prior arts did not address the issue of identifying vehicles with similar physical features from multiplexed snapshots or images. Also, the prior arts did not address the issue of analyzing motor vehicles that may be present in interlaced images.
Therefore, it is an aim of this present invention to address the aforesaid technical disadvantages by introducing a system and method for extracting objects from interlaced images that is capable of identifying similar motor vehicles moving sequentially in the same direction along the same path and similar motor vehicles moving sequentially in different directions along the same path by grouping and categorizing similar motor vehicles for analysis of the interlaced images.
SUMMARY OF THE PRESENT INVENTION
The present invention relates to a system and method for extracting objects from interlaced images. The system as disclosed in one of the embodiment of the present invention comprises a video image acquisition module that captures images of a moving object along a path of interest using optical sensors a moving object detection module that differentiates movement of the object from a static background of the images captured a visualization display that shows the captured images of the object, which the system is characterized by a calibration data profiling module that represents connections of the optical sensors with each other and the paths to be taken by the object and derives the mean of the travel time of the object based on recorded event time of the object detected, and a delayed real-time detection module that calculates the time taken by the
object to travel along the said path based on real-time delayed detection, and compares the said time with the mean of the travel time of the object based on the recorded event time of the object detected. The method as disclosed in a further embodiment of the present invention comprises the steps of generating a path topology map to represent connections of optical sensors with each other, computing a calibrated event profile module to derive the mean of travel time of the object based on time records of the object detected, acquiring video images of the object moving along the paths, detecting moving object, deriving a travel time of the object along each of the path based on real-time delayed detection, comparing the travel time of the object along each of the paths based on real-time delayed detection with the travel time of the object based on means of the calibrated event profile, extracting and grouping the video images of the objects into categories for display, and displaying a timeline of the images of the object based on the categories.
It is an object of the present invention to provide a system and method that is capable of extracting objects or similar objects from interlaced images.
It is another object of the present invention to provide a system and method that is capable of extracting vehicle or similar vehicles from interlaced images.
It is a further object of the present invention to provide a system and method that is capable of extracting vehicle or similar vehicles passing along a path of interest from interlaced images.
It is still an object of the present invention to provide a system and method for monitoring movements of a vehicle or similar vehicles as part of traffic flow monitoring.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 illustrates the flow diagram of the method for extracting objects from interlaced images.
Figure 2 illustrates the flow diagram of the step of generating a path topology map.
Figure 3 illustrates the generation of a path topology map and a look-up table for the path topology map. Figure 4 illustrates the flow diagram of the step of computing a calibrated event profile module.
Figure 5 illustrates the calibrated event profile for each path in the path topology map.
Figure 6 illustrates the calibrated travel time for each respective path in the path topology map.
Figure 7 illustrates the mean of calibrated travel time for each path.
Figure 8 illustrates the flow diagram of the step of acquiring video images of the object moving along the paths.
Figure 9 illustrates the extraction of video streams into frames.
Figure 10 illustrates the flow diagram of the step of detecting the presence of moving objects.
Figure 11 illustrates the scanning of the frames to differentiate a moving object from background image.
Figure 12 illustrates the flow diagram of the step of deriving the travel time of the object along each of the path based on real-time delayed detection.
Figure 13 illustrates the computation of the travel time of the moving object.
Figure 14 illustrates the flow diagram of the step of comparing the travel time of the object.
Figure 15 illustrates the computation of the value of the difference in percentage (%) and the comparison of the value with the threshold having a range of ±2.5%.
Figure 16 illustrates the flow diagram of the step of extracting and grouping the images of the objects into categories for display. Figure 17 illustrates the storing of images in display buffers.
Figure 18 illustrates the flow diagram of the step of displaying the timeline of the images of the object based on the categories. Figure 19 illustrates the display of the images in a timeline.
DETAILED DESCRIPTION OF THE PRESENT INVENTION
The above mentioned and other features and objects of this invention will become more apparent and better understood by reference to the following detailed description. It should be understood that the detailed description made known below is not intended to be exhaustive or limit the invention to the precise form disclosed as the invention may assume various alternative forms. On the contrary, the detailed description covers all the relevant modifications and alterations made to the present invention, unless the claims expressly state otherwise.
One of the embodiment of the present invention illustrates a system for extracting objects from interlaced images comprising of a video image acquisition module that captures images of a moving object along a path of interest using optical sensors a moving object detection module that differentiates movement of the object from a static background of the images captured a visualization display that shows the captured images of the object, which is characterized by a calibration data profiling module that represents connections of the optical sensors with each other and the paths to be taken by the object and derives the mean of the travel time of the object based on recorded event time of the object detected, and a delayed real-time detection module that calculates the time taken by the object to travel along the said path based on real-time delayed detection, and compares the said time with the mean of the travel time of the object based on the recorded event time of the object detected.
According to the embodiment, the calibration data profiling module generates a path topology map to represent the connections of the optical sensors with each other and the paths to be taken by the object. It is disclosed in the present embodiment of the present invention that the mean of travel time is changeable to any time period of the day and week.
A further embodiment of the present invention illustrates a method for extracting objects from interlaced images comprising the steps of generating a path topology map to represent connections of optical sensors with each other, computing a calibrated event profile module to derive the mean of travel time of the object based on time records of the object detected, acquiring video images of the object moving along the paths, detecting moving object, deriving a travel time of the object along each of the path based on real-time delayed detection, comparing the travel time of the object along each of the paths based on real- time delayed detection with the travel time of the object based on means of the calibrated event profile, extracting and grouping the video images of the objects into categories for display, and displaying a timeline of the images of the object
based on the categories. The categories mentioned herein are determined based on similarities of the images and the images are grouped into their respective categories based on the similarities that occur in the images. The steps described herein are illustrated in Figure 1.
Further to the method disclosed herein, the step of generating the path topology map further comprises the steps of retrieving index numbers of the optical sensors and respective location of the optical sensors within the area of interest from a first database, setting a path to any two of the optical sensors that are connected to each other, generating a path topology map to represent the said paths, transforming the path topology map into a look-up table, and storing the path topology map and the look-up table. In one of the embodiments of the invention, the optical sensor is identified as CamlD. If one of the optical sensor is named CamA and the other optical sensor is named CamB, the path is set with either CamA_CamB and CamB_CamA. The steps described herein are illustrated in Figure 2. The path topology map and look-up table are illustrated in Figure 3.
The second step of the method, which is the computation of the calibrated event profile module further comprising the steps of retrieving a period of time of the object's recorded timestamp for each optical sensor from a second database, searching for paths that are connected to the optical sensors from the path topology map, computing a travel time of the object for each path in a calibrated event profile module, calculating a mean of calibrated travel times for each path, and storing the mean calibrated travel time into a third database. During the retrieval of timestamp, all information related to the particular optical sensor is also retrieved. From the path topology map, the paths that are connected to the particular optical sensor is determined when the path of CamA_CamB shows TRUE or the path of CamB_CamA shows TRUE. Figure 5 illustrates the calibrated event profile for each path in the path topology map and Figure 6 illustrates the calibrated travel time for each respective path in the path topology map. In a preferred embodiment of the invention, the calibration method is
carried with constraints such that the object's speed is within a speed limit, the object is not allowed to stop midway of the path, the rear part of the object is not allowed to overtake the front part of the object along the path, and the path is a single lane path. Upon computing a travel time for each path, the computation switches to a subsequent optical sensor having the object's timestamp. If the subsequent optical sensor does not have any timestamp, the computation proceeds with the calculation of the mean of calibrated travel times for each path. Figure 7 illustrates the mean of calibrated travel time for each path. Before storing the mean of calibrated travel times for each path, the computation also determines if a subsequent path exists. If the subsequent path exists, the steps of retrieving timestamps and information of the optical sensors and the following steps are repeated until no subsequent path exists. The steps described herein are illustrated in Figure 4. Subsequently, video images are acquired and the step of acquiring the video images further comprises the steps of retrieving index numbers of the optical sensors and locations within the area of interest from a first database, determining video stream availability from the optical sensors, extracting the video streams into frames, and storing the frames into video buffers. Figure 8 illustrates the flow diagram of the steps described herein and Figure 9 illustrates the extraction of video streams into frames.
Thereafter, presence of moving objects is determined. The step of detecting the presence of moving objects further comprises the steps of retrieving frames from video buffers according to the index numbers of the optical sensors, scanning the frames to differentiate a moving object from background image, extracting frames having the moving object together with a recorded timestamp and index number of the optical sensor, and updating the recorded timestamp and index number in a camera event timestamp buffer. However, if the current optical sensor's index has been detected and recorded previously, the previous detected and recorded time with the last value is overwritten. The steps
described herein are illustrated in Figure 10 and the step of scanning the frames to differentiate a moving object from background image is illustrated in Figure 1 1.
The step of deriving the travel time of the object along each of the path based on real-time delayed detection further comprises the steps of retrieving an index number and recorded timestamp of a first optical sensor that is connected to one end of a path from a camera event timestamp buffer, retrieving an index number and recorded timestamp a second optical sensor that is connected to the other end of the said path from the camera event timestamp buffer, computing a difference between the recorded timestamps from the first optical sensor and the second optical sensor to obtain a travel time of the moving object, and storing the travel time of the moving object into a fourth database. Figure 12 illustrates the steps described herein and Figure 13 illustrates the computation of the travel time of the moving object.
Then, the travel times of the object are compared by retrieving information of the optical sensor where the moving object is detected, retrieving mean calibrated travel time from a third database, retrieving travel time of the moving object based on real-time delayed detection from a fourth database, computing differences between the travel time obtained from real-time delayed detection and the mean calibrated travel time to determine whether the object is the same object or a different object, determining whether the computed difference has a value that falls within a threshold, and determining validity of the value of the difference. The value of the difference computed is in percentage (%) and the value is compared with the threshold having a range of ±2.5%. The validity of the said value is determined based on whether the value falls within this range. Figure 14 illustrates the steps described herein and Figure 15 illustrates the computation of the value of the difference in percentage (%) and the comparison of the value with the threshold having a range of ±2.5%.
Next, the video images of the objects are extracted and grouped into similar categories for display. The step of extracting and grouping video images of the
objects further comprises the steps of determining whether the same object is detected, obtaining information of the optical sensor, obtaining images of the detected object based on the real-time delayed detection using the recorded timestamp, determining whether category of the object exists, and storing the information of the optical sensor and the images of the object into a first display buffer when the category of the object exists, wherein the information of the optical sensor and the images of the object is stored into a second display buffer when the category does not exists. Figure 16 illustrates the steps described herein and Figure 17 illustrates the storing of images in display buffers.
Lastly, the step of displaying the timeline of the video images of the object based on the categories further comprises the steps of determining whether there is any update from display buffers, obtaining sequential images of the event detection and counter checking the sequential images with the information of the recorded time for the arrangement of the images in the order of time, sending the images to a display monitor, determining whether a display buffer is generated, and repeating the step of obtaining and counter checking of the sequential images and sending the images to the display monitor. The arrangement of the images is in the order of time forming a timeline and the timeline is arranged in sequence and categories of the images. Figure 18 illustrates the steps described herein and Figure 19 illustrates the display of the images in a timeline.
Further embodiment of the present invention discloses that the object as described is a vehicle, which the system and method may be used for detecting a vehicle or similar vehicles in interlaced images.
The invention described herein is susceptible to variations, modifications and/or additions other than those specifically described and it is to be understood that the invention includes all such variations, modifications and/or additions which fall within the scope of the following claims.
Claims
1. A system for extracting objects from interlaced images comprising of: a video image acquisition module that captures images of a moving object along a path of interest using optical sensors;
a moving object detection module that differentiates movement of the object from a static background of the images captured;
a visualization display that shows the captured images of the object;
is characterized by
a calibration data profiling module that represents connections of the optical sensors with each other and the paths to be taken by the object and derives the mean of the travel time of the object based on recorded event time of the object detected; and
a delayed real-time detection module that calculates the time taken by the object to travel along the said path based on real-time delayed detection, and compares the said time with the mean of the travel time of the object based on the recorded event time of the object detected.
2. A system according to claim 1 , wherein the calibration data profiling module generates a path topology map to represent the connections of the optical sensors with each other and the paths to be taken by the object.
3. A method for extracting objects from interlaced images comprising the steps of:
generating a path topology map to represent connections of optical sensors with each other;
computing a calibrated event profile module to derive the mean of travel time of the object based on time records of the object detected;
acquiring video images of the object moving along the paths;
detecting moving object;
deriving a travel time of the object along each of the path based on realtime delayed detection;
comparing the travel time of the object along each of the paths based on real-time delayed detection with the travel time of the object based on means of the calibrated event profile;
extracting and grouping the images of the objects into categories for display; and
displaying a timeline of the images of the object based on the categories.
4. The method according to claim 3, wherein the step of generating a path topology map further comprises the steps of:
retrieving index numbers of the optical sensors and respective location of the optical sensors within the area of interest from a first database;
setting a path to any two of the optical sensors that are connected to each other;
generating a path topology map to represent the said paths;
transforming the path topology map into a look-up table; and
storing the path topology map and the look-up table.
5. The method according to claim 3, wherein the step of computing a calibrated event profile module further comprising the steps of:
retrieving a period of time of the object's recorded event time for each optical sensor from a second database;
searching for paths that are connected to the optical sensors from the path topology map;
computing a travel time of the object for each path;
calculating a mean of calibrated travel times for each path; and storing the mean calibrated travel time as a calibrated event profile into a third database.
6. The method according to claim 3, wherein the step of acquiring video images further comprises the steps of:
retrieving index numbers of the optical sensors and locations within the area of interest from a first database;
determining video stream availability from the optical sensors;
extracting the video streams into frames; and
storing the frames into video buffers.
7. The method according to claim 3, wherein the step of detecting moving object further comprises the steps of:
retrieving frames from video buffers according to the index numbers of the optical sensors;
scanning the frames to differentiate a moving object from background image;
extracting frames having the moving object together with a recorded timestamp and index number of the optical sensor; and
updating the recorded timestamp and index number in a camera event timestamp buffer.
8. A method according to claim 3, wherein the step of deriving a travel time of the object along each of the path based on real-time delayed detection, further comprises the steps of:
retrieving an index number and recorded event time of a first optical sensor that is connected to one end of a path from a camera event timestamp buffer;
retrieving an index number and recorded event time a second optical sensor that is connected to the other end of the said path from the camera event timestamp buffer;
computing a difference between the recorded event times from the first optical sensor and the second optical sensor to obtain a travel time of the moving object; and
storing the travel time of the moving object into a fourth database.
9. The method according to claim 3, wherein the step of comparing the travel times of the moving object, further comprises the steps of:
retrieving information of the optical sensor where the moving object is detected;
retrieving mean calibrated travel time from a third database;
retrieving travel time of the moving object based on real-time delayed detection from a fourth database;
computing differences between the travel time obtained from real-time delayed detection and the mean calibrated travel time to determine whether the object is the same object or a different object;
determining whether the computed difference has a value that falls within a threshold; and
determining validity of the value of the difference.
10. A method according to claim 3, wherein the step of extracting and grouping the images of the objects into similar category for display, further comprises the steps of:
determining whether the same object is detected;
obtaining information of the optical sensor;
obtaining images of the detected object based on the real-time delayed detection using the recorded event time;
determining whether category of the object exists; and
storing the information of the optical sensor and the images of the object into a first display buffer when the category of the object exists;
wherein the information of the optical sensor and the images of the object is stored into a second display buffer when the category does not exists.
1 1. A method according to claim 3, wherein the step of displaying a timeline of the images of the object based on the categories further comprises the steps of:
determining whether there is any update from a first display buffer;
obtaining images of the event detection in sequence and counter checking the images with the information of the recorded time for the arrangement of the images in the order of time;
sending the images to a display monitor; and
determining whether a second display buffer is generated,
wherein the step of obtaining and counter checking of the images and sending the images to the display monitor are repeated when a second display buffer is generated.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| MYPI2013702352A MY177608A (en) | 2013-12-04 | 2013-12-04 | A system and method for extracting objects from interlaced images |
| MYPI2013702352 | 2013-12-04 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2015084148A1 true WO2015084148A1 (en) | 2015-06-11 |
Family
ID=51663414
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/MY2014/000162 Ceased WO2015084148A1 (en) | 2013-12-04 | 2014-06-05 | A system and method for extracting objects from interlaced images |
Country Status (2)
| Country | Link |
|---|---|
| MY (1) | MY177608A (en) |
| WO (1) | WO2015084148A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12205376B2 (en) | 2014-09-26 | 2025-01-21 | Nec Corporation | Object tracking apparatus, object tracking system, object tracking method, display control device, object detection device, and computer-readable medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6240217B1 (en) | 1997-02-24 | 2001-05-29 | Redflex Traffic Systems Pty Ltd | Digital image processing |
| US20130050492A1 (en) | 2011-08-26 | 2013-02-28 | Michael Lehning | Method and Apparatus for Identifying Motor Vehicles for Monitoring Traffic |
-
2013
- 2013-12-04 MY MYPI2013702352A patent/MY177608A/en unknown
-
2014
- 2014-06-05 WO PCT/MY2014/000162 patent/WO2015084148A1/en not_active Ceased
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6240217B1 (en) | 1997-02-24 | 2001-05-29 | Redflex Traffic Systems Pty Ltd | Digital image processing |
| US20130050492A1 (en) | 2011-08-26 | 2013-02-28 | Michael Lehning | Method and Apparatus for Identifying Motor Vehicles for Monitoring Traffic |
Non-Patent Citations (4)
| Title |
|---|
| ELLIS T ET AL: "Active video-based surveillance system", IEEE SIGNAL PROCESSING MAGAZINE, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 22, no. 2, 1 March 2005 (2005-03-01), pages 25 - 37, XP011128091, ISSN: 1053-5888, DOI: 10.1109/MSP.2005.1406473 * |
| LILIANA LO PRESTI ET AL: "Path Modeling and Retrieval in Distributed Video Surveillance Databases", IEEE TRANSACTIONS ON MULTIMEDIA, IEEE SERVICE CENTER, PISCATAWAY, NJ, US, vol. 14, no. 2, 1 April 2012 (2012-04-01), pages 346 - 360, XP011436293, ISSN: 1520-9210, DOI: 10.1109/TMM.2011.2173323 * |
| WANG XIAOGANG ED - CALDERARA SIMONE ET AL: "Intelligent multi-camera video surveillance: A review", PATTERN RECOGNITION LETTERS, vol. 34, no. 1, 20 July 2012 (2012-07-20), pages 3 - 19, XP028955937, ISSN: 0167-8655, DOI: 10.1016/J.PATREC.2012.07.005 * |
| ZULAIKHA KADIM ET AL: "Video analytics algorithm for detecting objects crossing lines in specific direction using blob-based analysis", ADVANCES IN MOBILE COMPUTING & MULTIMEDIA, ACM, 2 PENN PLAZA, SUITE 701 NEW YORK NY 10121-0701 USA, 3 December 2012 (2012-12-03), pages 306 - 312, XP058029720, ISBN: 978-1-4503-1307-0, DOI: 10.1145/2428955.2429016 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12205376B2 (en) | 2014-09-26 | 2025-01-21 | Nec Corporation | Object tracking apparatus, object tracking system, object tracking method, display control device, object detection device, and computer-readable medium |
| US12307776B2 (en) * | 2014-09-26 | 2025-05-20 | Nec Corporation | Object tracking apparatus, object tracking system, object tracking method, display control device, object detection device, and computer-readable medium |
| US12380699B2 (en) | 2014-09-26 | 2025-08-05 | Nec Corporation | Object tracking apparatus, object tracking system, object tracking method, display control device, object detection device, and computer-readable medium |
| US12430920B2 (en) | 2014-09-26 | 2025-09-30 | Nec Corporation | Object tracking apparatus, object tracking system, object tracking method, display control device, object detection device, and computer-readable medium |
Also Published As
| Publication number | Publication date |
|---|---|
| MY177608A (en) | 2020-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325089B (en) | Method and apparatus for tracking object | |
| CN109887281B (en) | Method and system for monitoring traffic incident | |
| US7580547B2 (en) | Electronic traffic monitor | |
| AU2009311052B2 (en) | Motion detection method, apparatus and system | |
| CN110738150B (en) | Camera linkage snapshot method and device and computer storage medium | |
| CN107590486B (en) | Moving object identification method and system, and bicycle flow statistical method and equipment | |
| KR20160062880A (en) | road traffic information management system for g using camera and radar | |
| CN106097480A (en) | Vehicle operation data record system | |
| CN105608209B (en) | Video annotation method and video annotation device | |
| CN110969142A (en) | A method for extracting abnormal driving scenes based on natural driving data of connected vehicles | |
| CN111310698A (en) | License plate recognition method and device and inspection vehicle | |
| CN112419364A (en) | Target tracking method and system based on image feature matching | |
| CN118691648A (en) | Cross-camera block pedestrian matching method and device based on spatiotemporal information | |
| Hadiuzzaman et al. | Pixel-based heterogeneous traffic measurement considering shadow and illumination variation | |
| Mostafa et al. | Automatic Vehicle Classification and Speed Tracking | |
| CN112991769A (en) | Traffic volume investigation method and device based on video | |
| JP2002367077A (en) | Traffic congestion determination device and traffic congestion determination method | |
| CN103473925A (en) | Verification method of road vehicle detection system | |
| CN109558770A (en) | True value mask method | |
| CN116794650A (en) | Millimeter wave radar and camera data fusion target detection method and device | |
| CN119181254B (en) | Evidence obtaining method, system, device, equipment and medium for solid line lane change behavior of vehicle | |
| WO2015084148A1 (en) | A system and method for extracting objects from interlaced images | |
| CN109146910B (en) | An evaluation method of video content analysis index based on target positioning | |
| WO2020101595A3 (en) | System and method for automatic calibration of vehicle position determining device in a traffic system | |
| JP2000149181A (en) | Traffic stream measurement system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14781684 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 14781684 Country of ref document: EP Kind code of ref document: A1 |