CN114022450A - Splicing effect judgment method for vehicle panoramic all-round inspection test - Google Patents
Splicing effect judgment method for vehicle panoramic all-round inspection test Download PDFInfo
- Publication number
- CN114022450A CN114022450A CN202111307476.6A CN202111307476A CN114022450A CN 114022450 A CN114022450 A CN 114022450A CN 202111307476 A CN202111307476 A CN 202111307476A CN 114022450 A CN114022450 A CN 114022450A
- Authority
- CN
- China
- Prior art keywords
- splicing
- panoramic
- dislocation
- vehicle
- gap
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims abstract description 39
- 230000000694 effects Effects 0.000 title claims abstract description 25
- 238000007689 inspection Methods 0.000 title description 2
- 238000001514 detection method Methods 0.000 claims abstract description 54
- 238000012545 processing Methods 0.000 claims abstract description 21
- 230000000007 visual effect Effects 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 description 29
- 238000012790 confirmation Methods 0.000 description 24
- 230000005856 abnormality Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000008450 motivation Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000005096 rolling process Methods 0.000 description 2
- 238000010998 test method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of vehicle testing, and particularly relates to a splicing effect judgment method for a vehicle panoramic all-round-looking test, which comprises the following steps of obtaining a splicing image of the panoramic all-round-looking test; step two, analyzing and processing the splicing map according to a preset detection rule to obtain a splicing test result; wherein, the detection rule includes: and detecting splicing gaps, splicing malpositions, splicing ghosts and splicing losses of the splicing map. The method and the device can improve the effectiveness of the image spliced by the panorama, and better assist and guide the driver.
Description
Technical Field
The invention belongs to the technical field of vehicle testing, and particularly relates to a splicing effect judgment method for a vehicle panoramic all-round-looking test.
Background
The vehicle panoramic all-round looking is a driving assistance system which is characterized in that 4 to 8 wide-angle cameras capable of covering all view field ranges around the vehicle are erected around the vehicle, and a plurality of paths of video images acquired at the same time are processed into a vehicle body top view of 360 degrees around the vehicle.
With the improvement of the requirements of people on vehicles, the panoramic view of the vehicles gradually becomes an important auxiliary means necessary for the vehicles. In the past, when a full-around vision system is not available, a driver of an automobile can only observe the conditions of two sides and the rear part through a rearview mirror when starting or parking, but due to the structure of an automobile body, the driver has a vision blind area, particularly a large passenger car and a large truck, and the area of the vision blind area is very large (wherein the area of the blind area around the automobile body of a large-scale commercial vehicle can reach 70 m)2) Accidents of rolling pedestrians and non-motor vehicles in the visual field blind area when the vehicles start or park frequently occur, and the accident risk of operating the vehicles in crowded places such as buses, urban logistics vehicles and the like is particularly serious.
In order to avoid the above situation as much as possible, a panoramic looking-around system is provided, so that a driver can know whether potential safety hazards exist around the driver when starting or parking. However, the development of the panoramic looking-around system needs to be perfect, and at present, there is no unified panoramic looking-around technical standard, which makes the test for the panoramic looking-around also very different.
At present, in a conventional panoramic all-round test, key point positions (such as point positions which need to be referred to during backing) on a splicing map of the panoramic all-round are generally detected, and the splicing effect is generally rough in judgment. Because most drivers now require panoramic looking around systems only to assist them in starting or parking. Therefore, in the test of the splicing effect, the images around the vehicle only need to be presented in one panoramic all-around spliced image, and a driver can complete vehicle operation by using the panoramic all-around spliced image, so that the detection of non-key positions such as the splicing position is very simple or even skipped.
However, as the demand of the driver for experience increases, the demand for panoramic looking around is no longer that the peripheral scene is basically restored, but a specific scene around the display is displayed with high quality, and the position of the peripheral object needs to be accurately displayed, which has a high demand for splicing the picture. On the other hand, if the splicing position of the spliced graph is abnormal, for example, the splicing gap is too large, a view blind area is also formed at the splicing gap, and if people or other dangerous objects exist in the splicing gap, potential safety hazards exist when a driver starts or parks the vehicle. However, since each manufacturer continues to use the previous production standards of each enterprise, it is only necessary to basically display the conditions around the vehicle body and to achieve sufficient definition of the critical positions, so that the basic requirements of the driver can be met.
Disclosure of Invention
The invention aims to provide a splicing effect judgment method for a vehicle panoramic test, which can improve the conformity/reducibility of panoramic spliced images and real conditions and better assist and guide a driver.
The basic scheme provided by the invention is as follows:
a splicing effect judgment method for a vehicle panoramic all-round test comprises the following steps:
step one, acquiring a splicing map of panoramic looking around;
step two, analyzing and processing the splicing map according to a preset detection rule to obtain a splicing test result; wherein, the detection rule includes: and detecting splicing gaps, splicing malpositions, splicing ghosts and splicing losses of the splicing map.
Basic scheme theory of operation and beneficial effect:
the mosaic of the panoramic surround view can be directly obtained from the display system of the panoramic surround view system. After the splicing map of the panoramic all-round view is acquired, the splicing map is compared with splicing gaps, splicing dislocation, splicing double images and splicing loss.
The splicing gap, namely the gap existing at the image splicing position when two images are spliced, if the gap is too large, a view blind area can be formed at the gap, especially for large vehicles, people or other dangerous objects can exist in the splicing gap, and potential safety hazards exist when a driver starts or parks. Splicing dislocation, namely when two images are spliced, the same places of the images are not aligned, if the splicing dislocation degree is higher, a driver needs to pay attention to the two dislocated image parts at the same time when checking the spliced images to start or park, and the use is very inconvenient; when the splicing misalignment is serious, the situation of position deformation even can occur, and the guide effect of the spliced image is caused. Splicing ghosts, namely ghosts existing at splicing slit positions, and the overlarge area or the large number of splicing ghosts can cause large interference on the visual field of a splicing position, so that the driver is interfered. The splicing loss, i.e. the incomplete place of the picture splicing position, is similar to the splicing gap, and if the lost area is too large, a view blind area is formed.
According to the method and the device, after the splicing gap, the splicing dislocation, the splicing ghost and the splicing loss are analyzed, the splicing test result is obtained. Whether the spliced part in the spliced picture exists or not can be comprehensively known, so that a driver can use the spliced picture inconveniently, and even a driver can have a potential safety hazard. The panoramic looking-around system of the vehicle successfully passes the test of the method, which shows that the spliced picture of the panoramic looking-around system can truly and completely show the surrounding situation of the vehicle, and a driver can directly know the abnormality existing in the current starting or parking through the whole picture, so that the use is more convenient and effective.
In conclusion, the method and the device can improve the conformity/reducibility between the panoramic spliced image and the real situation, and better assist and guide the driver.
Further, in the second step, when the splicing gap is detected, processing the data of the splicing gap part of the splicing image, generating a difference value according to the contrast parameter, and judging whether the splicing gap reaches the gap standard.
Has the advantages that: by the method, whether the gap exists and the specific data of the gap when the gap exists can be accurately known, so that whether the gap standard is met or not and where the gap standard is not met can be conveniently and visually known.
Further, in the second step, the standard of the gaps is that in the panoramic spliced image, the pixel area occupied by the spliced gaps is not more than 2% of the whole image.
Has the advantages that: 2% of the pixel area of the whole picture corresponds to the minimum area of a 5 year old child crouching down. Due to the arrangement, the size of the splicing gap on the tested splicing diagram can be ensured to be very small, and the situation that the risk is increased when starting or parking due to the fact that people or other dangerous objects exist in the gap can be avoided.
Further, in the second step, when the splicing dislocation is detected, the part with the dislocation is analyzed, and whether the splicing dislocation meets the dislocation standard or not is judged according to a preset difference threshold value.
Has the advantages that: through the mode, whether splicing dislocation exists or not can be accurately and visually known, and whether the place with the splicing dislocation is too large or not can be accurately known, so that whether the splicing dislocation reaches the preset dislocation standard or not can be accurately known.
Further, in the second step, the misalignment standard is that in the panoramic spliced image, the splicing misalignment amount along the vehicle length direction is not more than 3% of the total length of the panoramic spliced image, the splicing misalignment amount along the vehicle width direction is not more than 3% of the total width of the panoramic spliced image, and the single splicing misalignment length at each position in the visual range is less than 0.3 m.
Has the advantages that: due to the arrangement, the consistency of the splicing map reaching the dislocation standard can be ensured, and a driver does not need to pay attention to objects at the dislocation part at all times when using the splicing map, so that the convenience of using the splicing map is ensured.
Further, in the second step, when splicing ghost detection is carried out, whether ghosting occurs is identified, and whether the splicing ghost reaches a ghost standard is judged.
Has the advantages that: by the mode, the test rod is used as a reference object, the number and the size of the double images can be accurately identified, whether the spliced double images meet the double image standard or not can be conveniently and accurately known, and if the spliced double images do not meet the double image standard, the specific needs of improvement are met.
Further, in the second step, the ghost image standard is that in the panoramic spliced image, the area of a single plane spliced ghost image of each splicing gap on the ground is smaller than 0.09m within a visible range2The number of double images of a single three-dimensional object on each splicing gap is not more than 1.
Has the advantages that: by means of the arrangement, the number and the area of the double images on the splicing map meeting the double image standard can be guaranteed not to interfere with the visual field of the splicing position, and therefore drivers can be guaranteed not to be interfered due to splicing of the double images.
Further, in the step one, the obtained splicing image of the panoramic all-round vision is generated after a test board is placed according to the preset requirement; and step two, when the splicing loss is detected, generating a data report according to the length of the scale disappeared on the test board, and judging whether the splicing loss meets the loss standard.
Has the advantages that: the test board is a reference object, so that the specific data of the splicing situation can be conveniently and accurately analyzed, and whether the data of splicing loss on the splicing map reaches the loss standard or not can be known.
Further, the loss standard is that the total area of the plane splicing loss of each splicing gap on the ground is not more than 0.7m within the visual range of the panoramic spliced image2The splice loss width of the test board is less than 0.5m below the height 1m from the ground.
Has the advantages that: by means of the arrangement, the area of the incomplete splicing part on the splicing diagram meeting the loss standard is small, and the situation that a driver has safety influence when starting or parking due to the fact that people or other objects exist in the splicing loss part is avoided.
Further, the detection sequence of splicing gaps, splicing misplacements, splicing ghosts and splicing losses is as follows: and firstly detecting splicing gaps, then detecting splicing double images or splicing dislocation, and finally detecting splicing loss.
Has the advantages that: the detection sequence can further improve the overall detection efficiency on the basis of ensuring the detection accuracy.
Drawings
Fig. 1 is a flowchart of a first embodiment of the present invention.
Detailed Description
The following is further detailed by the specific embodiments:
example one
In the prior art, all manufacturers continue to use the test standard before splicing, namely the condition of the periphery of the vehicle body can be basically displayed, and the key position can be clearly displayed, so that the basic requirement of a driver can be met, and therefore, a person skilled in the art has no motivation to improve the detection mode of the spliced graph of the panoramic all-around system.
However, when a driver actually drives a vehicle, there are often problems in terms of a stitching gap, a stitching misalignment, a stitching ghost, a stitching loss, and the like, and therefore, when a stitching map in a panoramic environment is used as a visual field aid during starting or parking, the operation is not very convenient. However, on the one hand, since each manufacturer detects the mosaic as long as the key position can be clearly shown, the skilled person lacks motivation for improvement of the test method; on the other hand, the test standards of each manufacturer are different from each other, and it is difficult to unify them. Therefore, the detection of the mosaic of panoramic views still lacks a high standard method. The applicant is taken as a technician of a third party organization, and creatively provides a splicing effect detection method of a splicing map based on a panoramic all-around system, so that the panoramic all-around splicing map can be ensured to more truly present the real-time situation around a vehicle, the authenticity and the conformity are improved, and a better panoramic all-around splicing map can be provided for a driver.
When the test site is built, the built contents comprise a light source, a marking and a reference object, and black and white grid test cards are paved on the ground of a test area. It should be noted that, when the light source is adjusted, in order to ensure the test effect, in a field with uniform illumination, the brightness of each area around the vehicle should be kept consistent, there is no over-bright or over-dark area in a certain area, and the brightness difference between the brightest part and the darkest part in the picture is not more than 20%; under the scene of forward, reverse and lateral single-side illumination, the brightness of the junction of the light-emitting region and the backlight region should be naturally transited, and sudden brightness change or color cast should not occur.
And then, after the vehicle is parked in the designated area, the splicing effect test of the splicing chart can be carried out according to the test method in the application. It should be noted that, in this embodiment, after the splicing map is obtained, the testing of the splicing effect is automatically performed in the test system software, because the processing efficiency, stability, and consistency of the system are better than those of a manual operation.
As shown in fig. 1, a stitching effect determination method for a vehicle panoramic all-round test includes:
step one, acquiring a splicing map of panoramic looking around;
it should be noted that, when the detected item is a splice loss, the obtained splice map of the panoramic all around view is a splice map generated after the test board is placed according to the preset requirement. Specifically, when the test board is placed, the test board is erected on the ground, the center of the bottom edge of the test board is arranged on the splicing gap, and the bottom edge of the test board is perpendicular to the splicing gap (or the tangent line of the splicing gap) and is placed on the splicing gap 2m away from the vehicle body and perpendicular to the splicing gap (or the tangent line of the splicing gap). Wherein the test board has a height of not less than 100cm and a width of not less than 100 cm. The test board is provided with a horizontal scale which has the same length as the width of the test board and is not less than 5cm and is alternated with color blocks of different colors at the position of every 20cm from the bottom end to the top end. The width of the single color block on the horizontal scale is 10 cm. The bottom color of the test board is obviously distinguished from the ground, and the bottom color of the test board is yellow in the embodiment; the horizontal scales are red and white blocks. The surface of the test board should be made of a matte material. In this example, the test plate was 105cm in height and 100cm in width; the horizontal level is 5cm and the width is 100 cm.
Step two, analyzing and processing the splicing map according to a preset detection rule to obtain a splicing test result; wherein, the detection rule includes: and detecting splicing gaps, splicing malpositions, splicing ghosts and splicing losses of the splicing map. After the detection is finished, the detection result can be displayed in a table form. Therefore, the detection result can be more conveniently and quickly known by the staff, and when the detection item which does not meet the requirement exists, the type of the detection item which meets the requirement and the reason for the non-satisfaction can be quickly known.
Specifically, when the splicing gap is detected, the data of the splicing gap part of the splicing map is processed, a difference value is generated according to the color contrast of the splicing part, and whether the splicing gap meets the gap standard is judged. The standard of the gaps is that in the panoramic spliced image, the pixel area occupied by the spliced gaps is not more than 2% of the whole image. In this embodiment, the color contrast is determined by that, under red and neutral gray backgrounds, the CIEDE2000 color difference value of the images on two sides adjacent to the splicing gap is not higher than 20.
In the prior art, a splicing gap detection method is to observe a panoramic annular view by naked eyes, for example, a triangular cone or other three-dimensional objects are placed on the ground and detour for a circle at a position 1m away from a vehicle body, when loss/dislocation/double images of the three-dimensional objects appear on the panoramic view, marking points A1, B1, C1 and D1 are made on the ground, and whether black or colored and other macroscopic fixed splicing gaps exist is judged. According to the detection method, the reference objects are manually placed, and the number of the reference objects needing to be visited is large, so that the reference objects need to be placed along the vehicle for a week. The detection result of the detection method is directly determined by the reference object, and the different placing positions of the reference object directly lead to different detection results; in addition, because the detected vehicle types are not the same, the specific placing positions of the reference objects are different when the vehicle types are different, and the problem that the reference objects are difficult to place accurately cannot be directly solved by setting a reference line. Therefore, the accuracy and consistency of the existing detection method are difficult to guarantee.
In the scheme, a reference object is not set any more during testing of the splicing gap, but a thought is directly converted, and the splicing gap possibly causing a driving accident to occur to a driver is assumed to exist, so that what the gap should be. Based on the premise hypothesis, the inventor creatively and directly provides the splicing gap testing method. The image is converted into a pixel proportion test by the splicing gap detection method, the minimum area of a 5-year-old child squat is calculated (most children under the age of 5 are held and pulled by parents and basically cannot walk at the place where the vehicle is parked or started), the area of the ground is converted into the pixel of the image, the pixel corresponding to the area is obtained through proportion conversion (namely, the pixel area occupied by the splicing gap is not more than 2% of the whole picture), and the vehicle detected by the method can effectively avoid rolling the child in a blind area in the actual driving process. Meanwhile, the color of the test board observed at the position where the gap appears is changed, and the CIEDE2000 color difference value of the images at two adjacent sides of the splicing gap is regulated to be not higher than 20, so that the fuzzy degree of the boundary of the splicing gap can be determined, and the accurate identification of the splicing gap is further ensured. Through splicing gap detection and gap standards, whether gaps exist and specific data of the gaps when the gaps exist can be accurately known, whether the gaps reach the gap standards or not and where the gaps do not meet the standards when the gaps do not reach the standards can be conveniently and visually known. The size of the splicing gap on the tested splicing diagram is ensured to be very small, and the situation that the risk is increased when starting or parking due to the fact that people or other dangerous objects exist in the gap is avoided.
When the splicing dislocation is detected, the part with dislocation is analyzed, and whether the splicing dislocation reaches the dislocation standard or not is judged according to the preset difference threshold value. The dislocation standard is that in the panoramic spliced image, the splicing dislocation amount along the vehicle length direction is not more than 3% of the total length of the panoramic spliced image, the splicing dislocation amount along the vehicle width direction is not more than 3% of the total width of the panoramic spliced image, and the single splicing dislocation length at each position in the visual range is less than 0.3 m. Compared with the existing splicing dislocation of observing the panoramic annular view by naked eyes, the detection method measures the dislocation pixels and the distances corresponding to the lengths of the black and white grids and the width direction of the vehicle on the spliced image, can effectively improve the detection precision of the splicing dislocation, avoids the minimum requirement that the length of a single dislocation is not greater than the standard when the multiple dislocations appear in the detection range in the existing detection method and standard, is inconvenient for a driver to use, and even has the condition that the wrong looking is missed in partial small areas during use. Through concatenation dislocation detection and dislocation standard, can be accurate and audio-visual know the concatenation dislocation exist to and whether the place that exists the concatenation dislocation is too big, thereby the understanding concatenation dislocation that can be accurate has reached predetermined dislocation standard. The continuity of the splicing chart meeting the dislocation standard is ensured, and a driver does not need to pay attention to objects at the dislocation part at all times when using the splicing chart, so that the convenience of using the splicing chart is ensured.
And when the splicing ghost detection is carried out, identifying whether a ghost occurs or not, and judging whether the splicing ghost reaches a ghost standard or not. Wherein the ghost image standard is that in the panoramic spliced image, the area of a single plane spliced ghost image of each splicing gap on the ground is less than 0.09m in a visible range2The number of double images of a single three-dimensional object on each splicing gap is not more than 1.
The splicing ghost detection method in the prior art is the same as the splicing gap detection method, and the detection standard is that no splicing ghost which is visible to naked eyes appears. In the scheme, the splicing ghost detection method and the judgment method quantize the test result, judge whether splicing ghosting occurs or not by analyzing the repeated region of the black and white grid boundary, calculate the area of a ghost part and specify the maximum ghost area (0.09 m)2) And the detection precision is improved, so that unqualified products with double images and large double image areas are screened out. The number and the area of the double images on the splicing map meeting the double image standard are ensured not to interfere with the visual field of the splicing position. Through splicing ghost detection and ghost standards, the number and size of ghosts can be accurately identified, so that whether the splicing ghosts meet the ghost standards or not can be accurately known and judged, and specific needs to be improved if the splicing ghosts do not meet the ghost standards. The number and the area of the double images on the splicing map meeting the double image standard are ensured not to interfere with the visual field at the splicing position, so that the driver is ensured not to be interfered by splicing the double images.
When the splicing loss is detected, a data report is generated according to the length of the scale which disappears on the test board, and whether the splicing loss meets the loss standard or not is judged. Wherein the loss standard is that the total area of the planar splicing loss of each splicing gap on the ground is not more than 0.7m within the visual range of the panoramic spliced image2The splice loss width of the test board is less than 0.5m below the height 1m from the ground. Through splicing loss detection and loss standard, concrete data of splicing situation can be accurately analyzed, and therefore concrete data of splicing situation can be knownWhether the data lost by the splice on the splice map meets the loss criterion. On the splicing chart which meets the loss standard, the area of the incomplete splicing part is small, and the situation that a driver has safety influence when starting or parking due to the fact that people or other objects exist in the splicing loss part can be avoided.
According to the method and the device, after the splicing gap, the splicing dislocation, the splicing ghost and the splicing loss are analyzed, the splicing test result is obtained. Whether the spliced part in the spliced picture exists or not can be comprehensively known, so that a driver can use the spliced picture inconveniently, and even a driver can have a potential safety hazard.
The panoramic looking-around system of the vehicle successfully passes the test of the method, which shows that the spliced picture of the panoramic looking-around system can truly and completely show the surrounding situation of the vehicle, and a driver can directly know the abnormality existing in the current starting or parking through the whole picture, so that the use is more convenient and effective.
Example two
Compared with the first embodiment, the difference is that in the second step of the present embodiment, the splice gap position of the splice map is also detected, and if the splice gap position coincides with the vehicle body inherent blind area, it is recorded that the splice gap position is abnormal.
Compared with other shooting areas, the splicing gap position is a boundary area between two cameras and is also an area which is most prone to small problems, and if the splicing gap position coincides with an inherent blind area (such as an A-column blind area) of a vehicle body, when small problems occur, a driver is difficult to effectively and temporarily avoid due to the fact that the splicing gap position is located in a sight line blind area. By such an arrangement, the above-described situation can be avoided.
In this embodiment, the test of the splicing gap is completed first, and the tester connects the test sample piece, the vehicle, and the software analysis device first, arranges the red-gray test board on the ground, sets the environment uniform illumination, obtains the splicing picture, and after the image processing analysis, after obtaining the splicing gap result, performs the test of splicing ghost image/splicing dislocation.
When the splicing double images are detected, the test board is replaced on the ground, the detected straight lines of the checkerboards can be directly sorted in different directions due to the fact that splicing gaps are obtained, the upper straight line, the lower straight line, the left straight line, the right straight line and the left straight line of each checkerboard are positioned according to the relation that the distances between the straight lines are close, and the splicing double images are determined according to the double image portions of the outlines of the checkerboards. When splicing dislocation is detected, the test board is replaced on the ground, the upper, lower, left and right straight lines of each square are positioned according to the close relation of the distances between the straight lines, the boundaries of two adjacent chequerboards are positioned according to the straight lines, pixel values of the straight lines of the two adjacent chequerboards in the X direction and the Y direction are calculated and used as the distances between the adjacent chequerboards, and therefore delta X and delta Y data are obtained.
And finally, carrying out splicing loss test.
The splicing seam is detected firstly, and then the detection result of the splicing seam is utilized to carry out corresponding detection setting for guidance, so that the testing of splicing double images and splicing dislocation can be completed quickly. And finally, testing the splicing loss. Through such detection sequence, the result of the preceding detection can be used as the guide of the subsequent detection, and the overall detection efficiency can be further improved on the basis of ensuring the detection accuracy.
EXAMPLE III
Different from the first embodiment, the first step and the second step in this embodiment are both performed at the processing end, in this embodiment, the processing end is an industrial PC, and in other embodiments, the processing end may also be a backend server.
The method also comprises a step of arranging a guide and a step of confirming an abnormal report:
arranging a guide step, wherein a processing end sends scene guide information to a working end of a worker according to a preset sequence; in this embodiment, the working end is a smart phone loaded with a corresponding APP;
an exception report confirming step, wherein when the test result is that an exception exists, the processing end judges whether a corresponding effective confirmation signal sent by the working end is received within a preset time length; if not, the processing terminal sends a reminding signal to the working terminal; when the working end sends an abnormal confirmation signal, the working end synchronously sends self positioning information; when the processing end receives the abnormal confirmation signal and the corresponding positioning information, judging whether the positioning information is a preset display end position or not; if so, the processing end marks the corresponding abnormal confirmation signal as a valid confirmation signal; if not, the processing end sends a re-confirmation signal and confirmation request information to the corresponding working end.
The specific implementation process is as follows:
in order to smoothly carry out the arrangement of the site by the staff, the method can send scene guide information to the working ends of the staff according to a preset sequence. For example, the corresponding requirements of the splicing gap site requirement, the splicing ghost site requirement, the splicing dislocation site requirement and the splicing loss site requirement are respectively sent to the staff before the start of each test content. Thus, even a worker with insufficient working experience can quickly and accurately perform scene arrangement.
On the other hand, according to actual tests, only a few vehicles have abnormal stitching graphs of the panoramic all-around system, but once the abnormal stitching graphs occur, the abnormal stitching graphs have influence on users who purchase the vehicles. Therefore, in order to avoid such a situation as much as possible, each test result with abnormality needs a clear feedback of the fact that the worker has known.
Therefore, the method also comprises an abnormal report confirming step. The conventional abnormal confirmation mode in the field is that a worker watches a display result in front of a display end, and clicks a confirmation key on the display end when the display result is abnormal. However, in this method, after the worker continues to work for a long time, the worker easily makes a subconscious click confirmation key, but the specific abnormal content is not clear.
In the step of confirming the abnormal report of the method, when the test result is that the abnormality exists, the processing end judges whether a corresponding effective confirmation signal sent by the working end is received within a preset time length. To generate a valid acknowledgment signal, two conditions need to be fulfilled: firstly, a worker sends an abnormal confirmation signal through a user side; second, the worker sends an anomaly confirmation signal at the designated presentation end location. (display end, i.e. display device showing abnormal test results.)
The first condition is to draw attention from the staff, and if the staff can directly operate on the display end, although the operation is more convenient, the staff can easily operate habitually, and the attention of the staff is not high easily. However, the worker must pass through the user side (e.g., the smart phone loaded with the corresponding APP in this embodiment), although the operation is relatively troublesome, the worker must not look at the display side any more, the efforts of the worker are completely different, and further the degree of attention of the worker to the abnormality is different, and the duration of use is also different (the time used by the method is much longer, and the worker can well know the specific abnormal content in this period of time), so that the degree of understanding of the worker to the abnormal information can be improved.
The second condition can ensure the validity of the worker sending the abnormal confirmation signal, because if the worker sends the abnormal confirmation signal and is not at the display end, the worker is indicated to be at other places, and it is needless to say that the worker is likely to send the abnormal confirmation signal without carefully looking at the abnormal content. Therefore, the method of sending the abnormal confirmation signal through the user side does not improve the accuracy of the abnormal confirmation signal, but provides a chance for a worker to speculate and get the wrong result (only sending the abnormal confirmation signal at other places). Through the second condition, the staff can be restrained, and the validity of the abnormal confirmation signal sent by the staff is ensured.
In this scheme, only if the two conditions are satisfied simultaneously, the processing end marks the corresponding abnormal acknowledgement signal as a valid acknowledgement signal. And if the second condition is not met, the processing end sends a re-confirmation signal and confirmation requirement information to the corresponding working end, so that the corresponding working personnel can resend the abnormal confirmation signal according to the requirement. And if the valid confirmation signal is not received within the preset time length, the processing end also sends a reminding signal to the working end, so that the working end can pay attention to check the detection result with abnormality.
Through the process, the method can ensure that the specific content of each abnormal detection result can be clearly known by the working personnel.
The foregoing is merely an example of the present invention, and common general knowledge in the field of known specific structures and characteristics is not described herein in any greater extent than that known in the art at the filing date or prior to the priority date of the application, so that those skilled in the art can now appreciate that all of the above-described techniques in this field and have the ability to apply routine experimentation before this date can be combined with one or more of the present teachings to complete and implement the present invention, and that certain typical known structures or known methods do not pose any impediments to the implementation of the present invention by those skilled in the art. It should be noted that, for those skilled in the art, without departing from the structure of the present invention, several changes and modifications can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicability of the patent. The scope of the claims of the present application shall be determined by the contents of the claims, and the description of the embodiments and the like in the specification shall be used to explain the contents of the claims.
Claims (10)
1. A splicing effect judgment method for a vehicle panoramic all-round test is characterized by comprising the following steps of:
step one, acquiring a splicing map of panoramic looking around;
step two, analyzing and processing the splicing map according to a preset detection rule to obtain a splicing test result; wherein, the detection rule includes: and detecting splicing gaps, splicing malpositions, splicing ghosts and splicing losses of the splicing map.
2. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 1, characterized in that: and step two, processing the data of the splicing gap part of the splicing image and generating a difference value according to the contrast parameter when the splicing gap is detected, and then judging whether the splicing gap reaches the gap standard.
3. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 2, characterized in that: in the second step, the standard of the gaps is that in the panoramic spliced image, the pixel area occupied by the spliced gaps is not more than 2% of the whole image.
4. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 1, characterized in that: and in the second step, when the splicing dislocation is detected, analyzing a part generating the dislocation, and judging whether the splicing dislocation meets the dislocation standard according to a preset difference threshold value.
5. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 4, characterized in that: in the second step, the dislocation standard is that in the panoramic spliced image, the splicing dislocation amount along the vehicle length direction is not more than 3% of the total length of the panoramic spliced image, the splicing dislocation amount along the vehicle width direction is not more than 3% of the total width of the panoramic spliced image, and the single splicing dislocation length at each position in the visual range is less than 0.3 m.
6. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 1, characterized in that: and step two, identifying whether a ghost image appears during splicing ghost image detection, and judging whether the splicing ghost image meets a ghost image standard.
7. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 6, characterized in that: in the second step, the ghost image standard is that in the panoramic spliced image, the area of a single plane spliced ghost image of each splicing gap on the ground is smaller than 0.09m in a visible range2The number of double images of a single three-dimensional object on each splicing gap is not more than 1.
8. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 1, characterized in that: in the first step, the obtained splicing picture of the panoramic all-round sight is generated after a test board is placed according to the preset requirement; and step two, when the splicing loss is detected, generating a data report according to the length of the scale disappeared on the test board, and judging whether the splicing loss meets the loss standard.
9. The stitching effect determination method for the vehicle panoramic all-round test according to claim 8, characterized in that: secondly, the loss standard is that the total area of the plane splicing loss of each splicing gap on the ground is not more than 0.7m within the visual range of the panoramic spliced image2The splice loss width of the test board is less than 0.5m below the height 1m from the ground.
10. The stitching effect judgment method for the vehicle panoramic all-round test according to claim 1, characterized in that: the detection sequence of splicing gaps, splicing malpositions, splicing double images and splicing loss is as follows: and firstly detecting splicing gaps, then detecting splicing double images or splicing dislocation, and finally detecting splicing loss.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111307476.6A CN114022450A (en) | 2021-11-05 | 2021-11-05 | Splicing effect judgment method for vehicle panoramic all-round inspection test |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111307476.6A CN114022450A (en) | 2021-11-05 | 2021-11-05 | Splicing effect judgment method for vehicle panoramic all-round inspection test |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114022450A true CN114022450A (en) | 2022-02-08 |
Family
ID=80061556
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111307476.6A Pending CN114022450A (en) | 2021-11-05 | 2021-11-05 | Splicing effect judgment method for vehicle panoramic all-round inspection test |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114022450A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115797167A (en) * | 2022-11-02 | 2023-03-14 | 慧铁科技有限公司 | Jigsaw puzzle method based on rail transit vehicle images |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160191795A1 (en) * | 2014-12-30 | 2016-06-30 | Alpine Electronics, Inc. | Method and system for presenting panoramic surround view in vehicle |
| CN111179168A (en) * | 2019-12-12 | 2020-05-19 | 郑州嘉晨电器有限公司 | Vehicle-mounted 360-degree panoramic all-around monitoring system and method |
| CN113194309A (en) * | 2021-06-02 | 2021-07-30 | 重庆渝微电子技术研究院有限公司 | Imaging quality evaluation system of 360-degree panoramic looking-around equipment |
| CN113191952A (en) * | 2021-06-02 | 2021-07-30 | 重庆渝微电子技术研究院有限公司 | Spliced image evaluation method for 360-degree panoramic looking-around equipment |
| CN113450257A (en) * | 2020-08-18 | 2021-09-28 | 杭州海康汽车软件有限公司 | Method and device for testing image splicing flaws |
| WO2021213326A1 (en) * | 2020-04-20 | 2021-10-28 | 长城汽车股份有限公司 | Method, apparatus and system for calibrating panoramic surround view system |
-
2021
- 2021-11-05 CN CN202111307476.6A patent/CN114022450A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160191795A1 (en) * | 2014-12-30 | 2016-06-30 | Alpine Electronics, Inc. | Method and system for presenting panoramic surround view in vehicle |
| CN111179168A (en) * | 2019-12-12 | 2020-05-19 | 郑州嘉晨电器有限公司 | Vehicle-mounted 360-degree panoramic all-around monitoring system and method |
| WO2021213326A1 (en) * | 2020-04-20 | 2021-10-28 | 长城汽车股份有限公司 | Method, apparatus and system for calibrating panoramic surround view system |
| CN113450257A (en) * | 2020-08-18 | 2021-09-28 | 杭州海康汽车软件有限公司 | Method and device for testing image splicing flaws |
| CN113194309A (en) * | 2021-06-02 | 2021-07-30 | 重庆渝微电子技术研究院有限公司 | Imaging quality evaluation system of 360-degree panoramic looking-around equipment |
| CN113191952A (en) * | 2021-06-02 | 2021-07-30 | 重庆渝微电子技术研究院有限公司 | Spliced image evaluation method for 360-degree panoramic looking-around equipment |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115797167A (en) * | 2022-11-02 | 2023-03-14 | 慧铁科技有限公司 | Jigsaw puzzle method based on rail transit vehicle images |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112965503B (en) | Multi-path camera fusion splicing method, device, equipment and storage medium | |
| US9105115B2 (en) | Display systems and methods for displaying enhanced vision and synthetic images | |
| CN110706282A (en) | Automatic calibration method and device for panoramic system, readable storage medium and electronic equipment | |
| US8605949B2 (en) | Vehicle-based imaging system function diagnosis and validation | |
| DE102017115318A1 (en) | Heads-up display for monitoring vehicle perception activity | |
| CN106651963B (en) | A kind of installation parameter scaling method of the vehicle-mounted camera for driving assistance system | |
| CN111052179B (en) | Vehicle device, calibration result determination system, calibration result determination method, and computer-readable storage medium storing program | |
| JP6938107B1 (en) | How to measure road surface damage | |
| KR101547415B1 (en) | Monitoring the close vicinity around a commercial vehicle | |
| CN112484743B (en) | Vehicle-mounted HUD fusion live-action navigation display method and system thereof | |
| CN113432889B (en) | Panoramic image parking evaluation method and device for vehicle, vehicle and storage medium | |
| US10996469B2 (en) | Method and apparatus for providing driving information of vehicle, and recording medium | |
| CN114022450A (en) | Splicing effect judgment method for vehicle panoramic all-round inspection test | |
| Adamshuk et al. | On the applicability of inverse perspective mapping for the forward distance estimation based on the HSV colormap | |
| JP7207916B2 (en) | In-vehicle device | |
| US20190322222A1 (en) | Image Display Control Device | |
| JP3448443B2 (en) | Vehicle periphery confirmation system | |
| CN101044390B (en) | Sight distance measuring device and uses | |
| DE102006051539A1 (en) | Method and device for aerial image-based environmental detection in motor vehicles | |
| CN114724104A (en) | Method, device, electronic equipment, system and medium for detecting visual recognition distance | |
| JP6855254B2 (en) | Image processing device, image processing system, and image processing method | |
| CN113212309A (en) | Remote driving end video display method and device and electronic equipment | |
| JPS6378299A (en) | Automobile having signal recognition equipment using image processing art | |
| CN116958246B (en) | Camera-based automobile auxiliary function detection method | |
| EP4318374A1 (en) | Image processing method, vision system, fault locating method and motor vehicle |
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 |