CN107255470B - Obstacle detection device - Google Patents
Obstacle detection device Download PDFInfo
- Publication number
- CN107255470B CN107255470B CN201710412721.7A CN201710412721A CN107255470B CN 107255470 B CN107255470 B CN 107255470B CN 201710412721 A CN201710412721 A CN 201710412721A CN 107255470 B CN107255470 B CN 107255470B
- Authority
- CN
- China
- Prior art keywords
- vertical edge
- obstacle
- class
- vertical
- module
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/36—Videogrammetry, i.e. electronic processing of video signals from a single source or from different sources to give parallax or range information
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides an obstacle detection device which comprises a storage unit, a quasi-vertical edge detection module, an intersection judgment module, a position judgment module and an obstacle detection module. The class vertical edge detection module receives an input image and performs a class vertical edge detection program on the input image to obtain a plurality of class vertical edges. The intersection judging module judges whether the class vertical edges intersect with a preset virtual ground horizontal line or not. If one of the class vertical edges intersects with the virtual ground horizontal line, the position judgment module judges whether one end of the class vertical edge intersecting with the virtual ground horizontal line is positioned in the detection area. If one end of one of the quasi-vertical edges intersecting the virtual ground level is located within the detection area, the obstacle detection module determines that the existence of an obstacle is detected and provides an alert.
Description
The invention is a divisional application of an invention patent application with an application number of 201410102894.5 and an invention name of 'obstacle detection device' proposed on 19/03/2014.
Technical Field
The present invention relates to an obstacle detection device, and more particularly, to an obstacle detection device based on an image processing technique.
Background
As the number of automobiles increases, the probability of road accidents increases year by year. Obviously, besides the continuous improvement of vehicle technology in the power section, the improvement of safety during driving is another subject to be paid attention. Research shows that if the driver can be warned before collision, the probability of road accidents can be greatly reduced. Therefore, a good and accurate obstacle detection and warning system is a very important link in the current vehicle safety system.
Generally, radar or ultrasonic systems are commonly used in current car safety systems, but the radar or ultrasonic systems may reduce the recognition rate due to the inability to recognize the environment, and the electromagnetic power of the radar or ultrasonic systems may also have an influence on the human body. On the other hand, with the progress of image processing and image sensing technology, it is more and more common to use images for obstacle detection. Many image-based safety warning systems alert the driver to maintain a safe distance from the vehicle by identifying obstacles in the image. However, for various image processing and analysis methods, the recognition accuracy or the time required for operation of the system performance of the image processing and analysis methods are different. Therefore, how to provide an obstacle detection system with high accuracy based on an image processing method is one of the issues that those skilled in the art are interested in.
Disclosure of Invention
Therefore, the invention provides an obstacle detection device, which can improve the accuracy of detecting and identifying obstacles through images so as to provide correct obstacle warning to a driver timely.
The invention provides an obstacle detection device which comprises a storage unit, a quasi-vertical edge detection module, an intersection judgment module, a position judgment module and an obstacle detection module. The storage unit stores at least one image. The class vertical edge detection module is coupled to the storage unit, receives an input image and performs a class vertical edge detection procedure on the input image to obtain a plurality of class vertical edges. The intersection judging module judges whether the class vertical edges intersect with a preset virtual ground horizontal line or not. If one of the class vertical edges intersects with the virtual ground horizontal line, the position judgment module judges whether one end of the class vertical edge intersecting with the virtual ground horizontal line is positioned in the detection area. If one end of one of the quasi-vertical edges intersecting the virtual ground level is located within the detection area, the obstacle detection module determines that the existence of an obstacle is detected and provides an alert.
In an embodiment of the invention, the vertical-like edge detection module includes a first direction edge detection module and a second direction edge detection module to obtain a plurality of first direction edges and a plurality of second direction edges. And the class vertical edge detection module acquires the class vertical edges according to the included angle relationship between the first direction edges and the second direction edges.
In an embodiment of the invention, the vertical edge detection module aggregates a plurality of class vertical edges close to each other.
In an embodiment of the invention, the first direction edge detection of the vertical-like edge detection module is horizontal edge detection, and the second direction edge detection is vertical edge detection.
In another aspect, the present invention provides an obstacle detection device, which includes a storage unit, a similar vertical edge detection module, a similar vertical edge tracking module, a slope change calculation module, and an obstacle detection module. The storage unit at least stores a video stream containing a plurality of images. The class vertical edge detection module is coupled to the storage unit, and receives input images of the video stream and performs a class vertical edge detection procedure on the input images to obtain a plurality of class vertical edges. The class-vertical edge tracking module is coupled to the class-vertical edge detection module and tracks the class-vertical edges for a tracking time period. The slope change calculation module compares the slope change values of each type of vertical edge in the tracking time and judges whether the slope change values are smaller than a threshold value. If one of the slope change values is smaller than the threshold value, the obstacle detection module judges that the obstacle exists and provides a warning.
In an embodiment of the invention, the quasi-vertical edge detection module includes a first direction edge detection module and a second direction edge detection module to obtain a plurality of first direction edges and a plurality of second direction edges. And the similar vertical edge detection module acquires a first similar vertical edge according to the included angle relationship between the first direction edges and the second direction edges.
In an embodiment of the invention, the vertical edge detection module aggregates a plurality of first-type vertical edges that are close to each other and have similar slopes.
In an embodiment of the invention, the first direction edge detection of the vertical-like edge detection module is horizontal edge detection, and the second direction edge detection is vertical edge detection.
As described above, in one embodiment of the present invention, the obstacle detection device performs the class-vertical-edge detection program on the input image to obtain the class-vertical edge. Further, the obstacle detecting device detects an edge close to a vertical line or a completely vertical line by edge detection of different directivities. Therefore, the obstacle detection is carried out by judging whether the similar vertical edge generated by the similar vertical detection intersects with the horizon or not, the accuracy of identifying the obstacle can be further improved, the probability of error judgment is reduced, and the driving safety of a driver is improved.
In order to make the aforementioned and other features and advantages of the invention more comprehensible, embodiments accompanied with figures are described in detail below.
Drawings
FIG. 1 is a schematic diagram of an obstacle detection system according to the present invention;
fig. 2 is a block diagram of an obstacle detection device according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating a method for detecting obstacles according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating an obstacle detection scenario according to an embodiment of the invention;
fig. 5 is a block diagram of an obstacle detection device according to another embodiment of the present invention;
FIG. 6 is a flowchart illustrating a method for detecting obstacles according to another embodiment of the present invention;
fig. 7A and 7B are schematic diagrams illustrating an obstacle detection scenario according to an embodiment of the invention;
fig. 8 is a block diagram of an obstacle detection apparatus according to another embodiment of the present invention;
fig. 9 is a flowchart illustrating an obstacle detection method according to yet another embodiment of the invention.
Description of reference numerals:
10: an obstacle detection system;
12: an image acquisition unit;
15. 100, 400, 700: an obstacle detection device;
110. 410, 710: a storage unit;
120. 420, 720: a class vertical edge detection module;
130. 730: a rendezvous judgment module;
140. 740: a position judgment module;
150. 450, 770: an obstacle detection module;
430. 750: a class vertical edge tracking module;
440. 760: a slope change calculation module;
img1, Img2, Img 3: an image;
m1: an object;
l1: a virtual ground level line;
e1, E2, E2 ', E3, E3': a quasi-vertical edge;
z1: detecting a region;
c1, C1 ', C2, C2': a unit area;
s201 to S204: each step of the obstacle detection method according to an embodiment of the present invention;
s501 to S504: each step of the obstacle detection method according to another embodiment of the present invention;
s801 to S806: the steps of the method for detecting an obstacle according to another embodiment of the present invention are described.
Detailed Description
Fig. 1 is a schematic diagram of an obstacle detection system according to the present invention. Referring to fig. 1, an obstacle detection system 10 includes an image acquisition unit 12 and an obstacle detection device 15. The obstacle detection system 10 is adapted to a vehicle on which an image acquisition unit 12 is provided for acquiring an image of an environment around the vehicle. For example, the image capturing unit 12 may be disposed in front of the vehicle, such as above a windshield of the vehicle, to capture images of the lanes in front of the vehicle. The image capturing unit 12 is, for example, an image sensor having a Charge Coupled Device (CCD) or a Complementary Metal-Oxide Semiconductor (CMOS) element, and is configured to capture an image of an environment around the vehicle. More specifically, the image acquiring unit 12 may be, for example, a driving recorder camera having a video recording device, or a digital camera having a photographing function, but the present invention is not limited to these embodiments.
The obstacle detection device 15 is an electronic device having an image processing function, and may be implemented as a computer, a vehicle computer, or other vehicle electronic devices. The obstacle detection device 15 is directly or indirectly connected to the image acquisition unit 12 to receive the images or video streams acquired by the image acquisition unit 12. Accordingly, the obstacle detection device 15 can perform obstacle detection and warning according to the images or video streams acquired by the image acquisition unit 12. In the embodiment of the present invention, the obstacle detecting device 15 performs detection of an obstacle based on edge information in an image. In particular, the obstacle detecting device 15 detects a quasi-vertical edge (Quai-vertical edge) in which the edge direction is close to the vertical direction in the image.
Therefore, based on the principle that the ground level line in the driver's sight line is blocked by the obstacle, in an embodiment of the present invention, the obstacle detection device 15 determines whether the quasi-vertical edge in the image passes through the virtual level line on the image, so as to know whether the obstacle exists around the vehicle. Furthermore, in another embodiment of the present invention, the obstacle detection device 15 may further distinguish the obstacle contour from the lane lines on the road by the slope change of the similar vertical edge in the video stream. To further illustrate how obstacle detection may be performed based on class-vertical edges obtained by class-vertical edge detection, the present invention is described below with reference to various embodiments
Fig. 2 is a block diagram of an obstacle detection device according to an embodiment of the invention. Referring to fig. 2, the obstacle detection apparatus 100 includes a storage unit 110, a quasi-vertical edge detection module 120, a rendezvous determination module 130, a position determination module 140, and an obstacle detection module 150. The storage unit 110 is, for example, a random access memory (random access memory), a Flash memory (Flash) or other memories, and is used for storing image data. The quasi-vertical edge detection module 120 is coupled to the storage unit 110 to read or receive at least one image stored in the storage unit.
On the other hand, the quasi-vertical edge detection module 120, the intersection determination module 130, the position determination module 140 and the obstacle detection module 150 may be implemented by software, hardware or a combination thereof, which is not limited herein. The software is, for example, original code, application software, a driver, or a software module or function dedicated to realizing a specific function, or the like. Examples of the hardware include a Central Processing Unit (CPU), a programmable controller, a Digital Signal Processor (DSP), and other programmable general-purpose or special-purpose microprocessors (microprocessors). For example, the quasi-vertical edge detection module 120, the intersection determination module 130, the position determination module 140, and the obstacle detection module 150 are, for example, computer programs or instructions, which can be loaded into a processor of the obstacle detection device 100 to perform the obstacle detection function.
Fig. 3 is a flowchart illustrating an obstacle detection method according to an embodiment of the invention. Referring to fig. 2 and fig. 3, in the present embodiment, the obstacle detection method can be implemented by the obstacle detection apparatus 100 in fig. 2, for example. The steps of the obstacle detection method of the present embodiment are described below with reference to various elements of the obstacle detection apparatus 100.
First, in step S201, the class vertical edge detection module 120 receives an input image and performs a class vertical edge detection procedure on the input image to obtain a plurality of class vertical edges. Further, the quasi-vertical edge detection is used to detect edges in the image with an edge direction close to or the same as the vertical direction, which is perpendicular to the horizon direction. Specifically, in the embodiment of the present invention, the included angle between the edge direction of the similar vertical edge and the vertical direction is smaller than an angle value, and the angle value may depend on the practical application, and the present invention is not limited thereto.
Further, in an embodiment, the class-vertical edge detection module 120 includes, for example, a first direction edge detection and a second direction edge detection to obtain a plurality of first direction edges and a plurality of second direction edges. And the class vertical edge detection module acquires the class vertical edges according to the included angle relationship between the first direction edges and the second direction edges. It should be noted that, since the directions corresponding to the first direction edge detection and the second direction edge detection are different but are known parameters, the quasi-vertical edge detection module 120 may analyze the directionality of each edge in the image based on the angle relationship between the first direction edge and the second direction edge, and further obtain the quasi-vertical edge according to the edge direction in the image. In other words, the similar vertical edge detection module 120 may utilize different masks for edge detection with different directivities when performing the edge detection procedure. However, the present invention is not particularly limited to the respective directions of the first direction edge detection and the second direction edge detection, and may be determined according to the actual application and the requirement.
For example, the first direction edge detection may be horizontal edge detection, while the second direction edge detection may be vertical edge detection. Therefore, in one embodiment, the vertical-like edge detection module 120 may calculate the horizontal edge value and the vertical edge value by using Sobel masks with different directivities. Therefore, the quasi-vertical edge detection module 120 can not only detect the edge in the image by the horizontal edge value and the vertical edge value, but also analyze the directionality of the edge by the ratio of the horizontal edge value and the vertical edge value. The quasi-vertical edge detection module 120 may detect quasi-vertical edges in the image by the directionality of the edges. However, the present invention is not limited to the above embodiments, and the algorithm for detecting the edge directionality is applicable to the present invention. For example, the quasi-vertical edge detection module 120 may also utilize Prewitt masks for quasi-vertical edge detection of the present embodiment.
Thereafter, in step S202, the intersection determination module 130 determines whether the class vertical edges intersect with a preset virtual ground horizontal line. The virtual ground level is a preset reference line in the image, and the position of the virtual ground level in the image can be set according to the actual application situation. In addition to directly determining whether the quasi-vertical edge passes through the virtual ground level line, the intersection determination module 130 defines an image section including the virtual ground level line, for example, with the virtual ground level line as a reference line. For example, a region within ten pixels of the vertical distance from the virtual ground horizontal line may be used as an image interval for detecting whether the quasi-vertical edge intersects the virtual ground horizontal line. If part of the quasi-vertical edge is located within the image interval, the intersection determination module 130 may determine that the quasi-vertical edge intersects the virtual ground horizontal line.
It is noted that the vertical edge detection module 120 may also aggregate (grouping) multiple class vertical edges that are close to each other before the intersection determination module 130 determines whether the class vertical edge intersects the virtual ground horizontal line. Specifically, the vertical edge detection module 120 may know whether one type of vertical edge is close to another type of vertical edge according to the positions of the types of vertical edges, and aggregate a plurality of first type of vertical edges that are close to each other. In short, the quasi-vertical edges of the same obstacle may be integrated or merged by this step, thereby reducing the computational effort of the intersection determination module 130 to determine whether the quasi-vertical edges intersect the virtual ground level.
If one of the first-type vertical edges intersects the virtual ground horizontal line, the position determination module 140 determines whether one end of one of the first-type vertical edges intersecting the virtual ground horizontal line is located within the detection area in step S203. Further, the position determining module 140 may determine whether one end of one of the similar vertical edges intersecting the virtual ground horizontal line is located in the detection area according to the position information of the similar vertical edge, which may further distinguish whether the object associated with the similar vertical edge intersecting the virtual ground horizontal line is an obstacle blocking the driving. Specifically, the position determination module 140 may further identify an obstacle close enough to the distance between the vehicles through the determination of step S203. Since an object that is too far away from the vehicle does not obstruct the travel of the vehicle, the position determination module 140 may exclude the possibility that an object that is too far away from the vehicle is an obstacle. That is, the obstacle detection device 100 can distinguish an obstacle from a distant object by determining whether one end of the quasi-vertical edge is located in the detection area based on the fact that the edge of the distant object in the image does not fall in the detection area as described in the present embodiment.
Then, if one end of one of the first-type vertical edges intersecting the virtual ground horizontal line is located within the detection area, the obstacle detection module 150 determines that the presence of an obstacle is detected and provides an alert in step S204. The warning provided by the obstacle detection module 150 is, for example, one or a combination of a prompt text, a sound, and a light, but is not limited thereto. The obstacle detection module 150 may change the presentation of the prompt alert according to the actual application requirements.
Fig. 4 is a schematic diagram illustrating an obstacle detection scenario according to an embodiment of the invention. Referring to fig. 4, it is assumed that the image Img1 is a front image captured by an image capturing unit on a vehicle, and the image Img1 has a property M1. First, the obstacle detection device performs a class-vertical edge detection program on the image Img 1. In the example shown in fig. 4, the obstacle detecting device may detect at least the class vertical edge E1, and determine that the class vertical edge E1 intersects the virtual ground horizontal line L1. As shown in fig. 4, the quasi-vertical edge E1 is an edge that is close to the vertical direction at an angle. Further, the obstacle detecting device also determines that one end of the quasi-vertical edge E1 falls within the detection region Z1. Accordingly, since the quasi-vertical edge E1 intersects the virtual ground horizontal line L1 and one end thereof falls within the detection region Z1, the obstacle detection apparatus will determine that there is an obstacle near the vehicle and provide a warning to the driver.
Fig. 5 is a block diagram of an obstacle detection device according to another embodiment of the invention. Referring to fig. 5, the obstacle detection apparatus 400 includes a storage unit 410, a class-vertical edge detection module 420, a class-vertical edge tracking module 430, a slope change calculation module 440, and an obstacle detection module 450. The memory cell 410 is similar to or the same as the memory cell 110 of the previous embodiment, and therefore, the description thereof is omitted.
On the other hand, in the present embodiment, the similar vertical edge detection module 410 is coupled to the storage unit 410 to receive the plurality of images in the video stream. The similar vertical edge tracking module 430 is coupled to the similar vertical edge detection module 410 to track edges across multiple images. The class-vertical edge detection module 420, the class-vertical edge tracking module 430, the slope change calculation module 440, and the obstacle detection module 450 may be implemented by software, hardware, or a combination thereof, without limitation.
The software is, for example, original code, application software, a driver, or a software module or function dedicated to realizing a specific function, or the like. Examples of the hardware include a Central Processing Unit (CPU), a programmable controller, a Digital Signal Processor (DSP), or other programmable general-purpose or special-purpose microprocessors (microprocessors). For example, the class vertical edge detection module 420, the class vertical edge tracking module 430, the slope change calculation module 440, and the obstacle detection module 450 are, for example, computer programs or instructions that may be loaded into a processor of the obstacle detection device 400 to perform the function of obstacle detection.
Fig. 6 is a flowchart illustrating an obstacle detection method according to another embodiment of the invention. Referring to fig. 5 and fig. 6, in the present embodiment, the obstacle detection method can be implemented by using the obstacle detection apparatus 400 in fig. 5, for example. The steps of the obstacle detecting method according to the present embodiment will be described below with reference to various elements of the obstacle detecting apparatus 400.
First, in step S501, the class vertical edge detection module 420 receives input images of a video stream and performs a class vertical edge detection procedure on the input images to obtain a plurality of class vertical edges. In brief, the quasi-vertical edge detection module 420 obtains a video stream from the storage unit 410, wherein the video stream is composed of a plurality of images that are captured consecutively in time. The similar vertical edge detection module 420 performs similar vertical edge detection on the images included in the video streams, thereby obtaining a plurality of similar vertical edges on each image. The detailed implementation method for the class-vertical edge detection module 420 to perform class-vertical edge detection is similar to that of the class-vertical edge detection module 120 of the foregoing embodiment, and is not described herein again. In an embodiment, the class vertical edge detection module 420 may perform class vertical edge detection on each image to obtain a class vertical edge image corresponding to each image, where each class vertical image includes multiple class vertical edges.
Next, in step S502, the class vertical edge tracking module 430 continuously tracks the class vertical edges for a tracking time period, respectively. In step S503, the slope change calculation module 440 compares the slope change values of each type of vertical edge during the tracking time and determines whether the slope change values are smaller than a threshold value. Further, to compare the class vertical edges corresponding to each other on different images, the class vertical edge tracking module 430 continuously tracks the class vertical edges for a tracking time period, respectively. On the other hand, the slope change calculating module 440 compares the slopes of the similar vertical edges corresponding to each other on different images, and determines whether the slope change value is smaller than the threshold value.
In one embodiment, the vertical-like edge tracking module 430 and the slope change calculation module 440 may track the vertical-like edges and compare the slopes of the corresponding vertical-like edges using a Histogram of Gradient angles (HOG). In detail, the similar vertical edge image can be divided into a plurality of unit areas (cells), and a gradient angle histogram of each unit area can be established by counting the characteristic values of each pixel point. Therefore, by comparing the gradient angle histograms of the unit regions, the similar vertical edge tracking module 430 may obtain unit regions representing the same scene on different images to obtain unit regions corresponding to each other on different images, thereby achieving the purpose of tracking the similar vertical edge.
The slope change calculation module 440 can determine whether the slope change value is smaller than the threshold value by comparing the statistics of the gradient angle histogram. Further, through statistics of the gradient angle histograms, the slope variation calculating module 440 may obtain an angle value corresponding to a highest statistical bin (most significant bin) in each gradient angle histogram. Therefore, by comparing the angle values corresponding to the highest statistical grooves of the corresponding unit regions, the slope change calculation module 440 can obtain the slope changes of the various vertical edges.
It should be noted that, for a lane line on a road, the edge of the lane line changes direction as the vehicle turns. On the contrary, the edge of the obstacle does not change its direction due to the turning of the vehicle, and thus the obstacle detecting apparatus can determine whether the obstacle is detected by analyzing and comparing the slope change value of the vertical edge of the class for a tracking time. In other words, the obstacle detection device can distinguish the three-dimensional object from the planar object in the image according to the slope change value of the quasi-vertical edge, wherein the planar object is not enough to block the vehicle from traveling. Accordingly, if one of the slope change values is smaller than the threshold value, in step S504, the obstacle detection module 450 determines that the existence of the obstacle is detected and provides an alert.
Fig. 7A and 7B are schematic diagrams illustrating an obstacle detection scenario according to an embodiment of the invention. Referring to fig. 7A and 7B, it is assumed that the video stream acquired by the image acquiring unit on the vehicle includes an image Img2 and an image Img3, and the vehicle acquires an image Img2 and an image Img3 in a turning state. In other words, images Img2 and Img3 are images of the front of the vehicle acquired by the image acquisition unit on the vehicle at different times, and the acquisition time of image Img2 is earlier than the acquisition time of image Img 3. As is clear from the scenes in the images Img2 and Img3, the vehicle is assumed to be turning right.
First, the obstacle detection device performs a similar vertical edge detection program for each of the images Img2 and Img 3. In the example shown in fig. 7A and 7B, the obstacle detecting device may detect at least the class-vertical edge E2 and the class-vertical edge E3 of the image Img2, and the obstacle detecting device may detect at least the class-vertical edge E2 'and the class-vertical edge E3' of the image Img 3. Further, based on the calculation of the HOG, the obstacle detecting apparatus can know that the unit region C1 of the image Img2 and the unit region C1 'of the image Img3 are in a corresponding relationship with each other, and can know that the unit region C2 of the image Img2 and the unit region C2' of the image Img3 are in a corresponding relationship with each other.
Then, by comparing the histogram of gradient direction of the unit region C1 with the histogram of gradient direction of the unit region C1 ', the obstacle detecting device can know whether the slope change value between the class vertical edge E3 and the class vertical edge E3' is greater than the threshold value. Similarly, by comparing the histogram of gradient direction of the unit region C2 with the histogram of gradient direction of the unit region C2 ', the obstacle detecting device can know whether the slope change value between the class vertical edge E2 and the class vertical edge E2' is greater than the threshold value.
In this example, class vertical edge E3 and class vertical edge E3 'are edges associated with front obstacles, and class vertical edge E2 and class vertical edge E2' are edges associated with lane lines. As shown in fig. 7A and 7B, when the vehicle is turning, the slope between the class vertical edge E3 and the class vertical edge E3' does not change much. Conversely, when the vehicle is turning, the slope between the class vertical edge E2 and the class vertical edge E2' changes very significantly. Accordingly, the obstacle detecting device regards only the similar vertical edge E3 and the similar vertical edge E3' as edges corresponding to the obstacle. That is, the obstacle detecting device may determine whether an obstacle is detected by comparing the slope change value of each type of vertical edge during the tracking time.
Fig. 8 is a block diagram of an obstacle detection device according to another embodiment of the present invention. Referring to fig. 8, the obstacle detection apparatus 700 includes a storage unit 710, a similar vertical edge detection module 720, a rendezvous determination module 730, a position determination module 740, a similar vertical edge tracking module 750, a slope change calculation module 760, and an obstacle detection module 770. The above elements are the same as or similar to those in the embodiments shown in fig. 2 and fig. 5, and a person skilled in the art can refer to the related descriptions of fig. 2 and fig. 5 and so on, and will not be described herein again.
Fig. 9 is a flowchart illustrating an obstacle detection method according to yet another embodiment of the invention. Referring to fig. 8 and fig. 9, in the present embodiment, the obstacle detection method can be implemented by using the obstacle detection apparatus 700 in fig. 8, for example. In addition, in the embodiment, the obstacle detection device can perform obstacle detection through a single image and multiple images in the video stream at the same time, so that the obstacle detection accuracy is improved. The following describes the steps of the obstacle detecting method according to the present embodiment with various elements of the obstacle detecting apparatus 700.
In step S801, the class vertical edge detection module 720 receives input images of the video stream and performs a class vertical edge detection procedure on the input images to obtain a plurality of class vertical edges. In step S802, the intersection determination module 730 determines whether the class vertical edges intersect with a preset virtual ground horizontal line. If the determination in step S802 is negative, in step S803, the class-vertical edge tracking module 750 continuously tracks the class-vertical edges for a tracking time period.
Then, in step S805, the slope change calculation module 760 compares the slope change values of each type of vertical edge during the tracking time, and determines whether the slope change values are smaller than a threshold. If the determination in step S805 is yes or the determination in step S802 is yes, in step S804, the position determining module 740 determines whether one end of one of the quasi-vertical edges intersecting the virtual ground horizontal line is located in the detection region, and determines whether one end of one of the quasi-vertical edges corresponding to the slope variation value smaller than the threshold value is located in the detection region. If the determination in step S804 is yes, in step S806, the obstacle detection module 770 determines that the obstacle is detected and provides an alarm. The details of the above steps S801 to S806 can be analogized with reference to the related descriptions of fig. 1 to 7, and are not described herein again.
In summary, the obstacle detection device in an embodiment of the invention performs obstacle detection by image processing. The obstacle detection device detects edges close to a vertical line but not completely vertical by edge detection with different directivities, and judges whether an obstacle exists or not by using information generated by the similar vertical edges. Therefore, by detecting the similar vertical edge, the misjudgment caused by excessively complicated edge information or specific edge characteristics can be avoided, and the accuracy of identifying the obstacle is improved. In addition, in the process of detecting the obstacle by using the similar vertical edge, the invention can also identify the three-dimensional object and the plane object in the image by the video streaming generated by the single photographic device, and compared with the three-dimensional vision technology of a plurality of photographic devices, the invention can save a large amount of cost and reduce the calculation amount.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (4)
1. An obstacle detection device comprising:
a storage unit at least storing a video stream including a plurality of images; and
a class vertical edge detection module, coupled to the storage unit, for receiving input images of the video stream, performing a class vertical edge detection procedure on the input images to obtain a plurality of class vertical edges,
the obstacle detection device is characterized by comprising:
a class vertical edge tracking module, coupled to the class vertical edge detection module, for continuously and respectively tracking the class vertical edges for a tracking time;
the slope change calculation module is used for comparing the slope change value of each class vertical edge in the tracking time and judging whether the slope change value is smaller than a threshold value or not; and
and the obstacle detection module judges that the existence of the obstacle is detected and provides a warning if one of the slope change values is smaller than the threshold value.
2. The obstacle detection device of claim 1, wherein the similar vertical edge detection module comprises a first direction edge detection module and a second direction edge detection module to obtain a plurality of first direction edges and a plurality of second direction edges, and obtains the similar vertical edge according to an included angle relationship between the first direction edges and the second direction edges.
3. The obstacle detection apparatus of claim 2, wherein the vertical-like edge detection module aggregates the vertical-like edges that are close to each other and have similar slopes.
4. The obstacle detection apparatus according to claim 3, wherein the first direction edge detection of the vertical-like edge detection module is horizontal edge detection and the second direction edge detection is vertical edge detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710412721.7A CN107255470B (en) | 2014-03-19 | 2014-03-19 | Obstacle detection device |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410102894.5A CN104931024B (en) | 2014-03-19 | 2014-03-19 | Obstacle detector |
| CN201710412721.7A CN107255470B (en) | 2014-03-19 | 2014-03-19 | Obstacle detection device |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410102894.5A Division CN104931024B (en) | 2014-03-19 | 2014-03-19 | Obstacle detector |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107255470A CN107255470A (en) | 2017-10-17 |
| CN107255470B true CN107255470B (en) | 2020-01-10 |
Family
ID=54118302
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410102894.5A Active CN104931024B (en) | 2014-03-19 | 2014-03-19 | Obstacle detector |
| CN201710412721.7A Active CN107255470B (en) | 2014-03-19 | 2014-03-19 | Obstacle detection device |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410102894.5A Active CN104931024B (en) | 2014-03-19 | 2014-03-19 | Obstacle detector |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN104931024B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3309707A1 (en) * | 2016-10-13 | 2018-04-18 | Conti Temic microelectronic GmbH | Device and method for gradient-based object recognition of an input image |
| CN112965474A (en) * | 2019-11-27 | 2021-06-15 | 深圳市大富科技股份有限公司 | AGV navigation control method and system and storage medium |
| CN111198385A (en) * | 2019-12-26 | 2020-05-26 | 北京旷视机器人技术有限公司 | Obstacle detection method, obstacle detection device, computer device, and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0961157A (en) * | 1995-08-25 | 1997-03-07 | Nippon Seiki Co Ltd | Distance-measuring device |
| JPH1097699A (en) * | 1996-09-19 | 1998-04-14 | Toyota Motor Corp | Obstacle detecting device for vehicle |
| CN1576123A (en) * | 2003-07-03 | 2005-02-09 | 黄保家 | Anticollision system for motor vehicle |
| CN1782668A (en) * | 2004-12-03 | 2006-06-07 | 曾俊元 | Method and device for preventing collison by video obstacle sensing |
| EP1722332A1 (en) * | 2005-05-13 | 2006-11-15 | Nissan Motor Company Limited | Apparatus and method for image processing |
| CN101436300A (en) * | 2008-12-10 | 2009-05-20 | 东软集团股份有限公司 | Method and apparatus for dividing barrier |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3267105B2 (en) * | 1995-06-28 | 2002-03-18 | トヨタ自動車株式会社 | Object recognition device |
| CN1987357B (en) * | 2006-12-26 | 2010-05-19 | 浙江工业大学 | Intelligent parking auxiliary device based on omnibearing computer sight |
| JP4410292B1 (en) * | 2008-10-20 | 2010-02-03 | 本田技研工業株式会社 | Vehicle periphery monitoring device |
| JP2011033594A (en) * | 2009-08-06 | 2011-02-17 | Panasonic Corp | Distance calculation device for vehicle |
-
2014
- 2014-03-19 CN CN201410102894.5A patent/CN104931024B/en active Active
- 2014-03-19 CN CN201710412721.7A patent/CN107255470B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0961157A (en) * | 1995-08-25 | 1997-03-07 | Nippon Seiki Co Ltd | Distance-measuring device |
| JPH1097699A (en) * | 1996-09-19 | 1998-04-14 | Toyota Motor Corp | Obstacle detecting device for vehicle |
| CN1576123A (en) * | 2003-07-03 | 2005-02-09 | 黄保家 | Anticollision system for motor vehicle |
| CN1782668A (en) * | 2004-12-03 | 2006-06-07 | 曾俊元 | Method and device for preventing collison by video obstacle sensing |
| EP1722332A1 (en) * | 2005-05-13 | 2006-11-15 | Nissan Motor Company Limited | Apparatus and method for image processing |
| CN101436300A (en) * | 2008-12-10 | 2009-05-20 | 东软集团股份有限公司 | Method and apparatus for dividing barrier |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107255470A (en) | 2017-10-17 |
| CN104931024A (en) | 2015-09-23 |
| CN104931024B (en) | 2017-07-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112528878B (en) | Method and device for detecting lane line, terminal equipment and readable storage medium | |
| US20200097756A1 (en) | Object detection device and object detection method | |
| EP3007099B1 (en) | Image recognition system for a vehicle and corresponding method | |
| US8520071B2 (en) | Boundary line detection system with improved detection-performance | |
| CN109766867B (en) | Vehicle running state determination method and device, computer equipment and storage medium | |
| CN102792314A (en) | Cross traffic collision alert system | |
| US10229505B2 (en) | Motion determination system and method thereof | |
| JP6038422B1 (en) | Vehicle determination device, vehicle determination method, and vehicle determination program | |
| US20130202155A1 (en) | Low-cost lane marker detection | |
| EP2610778A1 (en) | Method of detecting an obstacle and driver assist system | |
| JP7251409B2 (en) | Lane change vehicle detection device, method and video surveillance device | |
| CN111213153A (en) | Target object motion state detection method, device and storage medium | |
| US9508000B2 (en) | Object recognition apparatus | |
| US20220055618A1 (en) | Apparatus, method, and computer program for object detection | |
| CN113297881A (en) | Target detection method and related device | |
| CN115331191B (en) | Vehicle type recognition method, device, system and storage medium | |
| CN112597846A (en) | Lane line detection method, lane line detection device, computer device, and storage medium | |
| CN107255470B (en) | Obstacle detection device | |
| US9245342B2 (en) | Obstacle detection device | |
| US9916672B2 (en) | Branching and merging determination apparatus | |
| US9824449B2 (en) | Object recognition and pedestrian alert apparatus for a vehicle | |
| Dai et al. | A driving assistance system with vision based vehicle detection techniques | |
| CN116152753A (en) | Vehicle information identification method and system, storage medium and electronic device | |
| JP7064400B2 (en) | Object detection device | |
| Banu et al. | Video based vehicle detection using morphological operation and hog feature extraction |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20211115 Address after: Taiwan Hsinchu Science Park, 12 Li Hang Road, China Patentee after: Huajing Technology Co., Ltd Address before: 3 / F, No. 12, Lixing Road, Hsinchu Patentee before: Nengjing Technology Co., Ltd |