CN111325789A - Curvature discontinuity point detection method based on discrete direction change sequence - Google Patents
Curvature discontinuity point detection method based on discrete direction change sequence Download PDFInfo
- Publication number
- CN111325789A CN111325789A CN202010077829.7A CN202010077829A CN111325789A CN 111325789 A CN111325789 A CN 111325789A CN 202010077829 A CN202010077829 A CN 202010077829A CN 111325789 A CN111325789 A CN 111325789A
- Authority
- CN
- China
- Prior art keywords
- sequence
- direction change
- point
- skeleton
- value
- 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.)
- Granted
Links
- 230000008859 change Effects 0.000 title claims abstract description 74
- 238000001514 detection method Methods 0.000 title abstract description 15
- 238000000034 method Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000009499 grossing Methods 0.000 claims abstract description 6
- 238000012545 processing Methods 0.000 claims description 21
- 230000004044 response Effects 0.000 claims description 12
- 239000012634 fragment Substances 0.000 claims description 4
- 238000009795 derivation Methods 0.000 abstract description 3
- 230000011218 segmentation Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000004069 differentiation Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 239000000178 monomer Substances 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research 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
- G06T7/70—Determining position or orientation of objects or cameras
-
- G06T5/70—
Abstract
The invention discloses a curvature discontinuity point detection method based on a discrete direction change sequence, which comprises the following steps: preprocessing, namely searching the end points of the acquired target edge skeleton and the existing Y-shaped connecting points, and removing the Y-shaped connecting points; obtaining an edge framework direction change sequence; the position of the V-shaped connection point is detected. The method has the advantages that a direction template is defined to express the direction change of the next pixel point to the current pixel point, so that the direction change sequence of the whole framework is obtained, and then proper smoothing and derivation operations are carried out on the sequence, so that the processed sequence can smooth the peak generated by noise and simultaneously highlight the peak generated by a V-shaped connection point, the automatic detection of curvature discontinuous points is realized, and an effective method is provided for the division of stacked groups in the image.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a curvature discontinuity point detection method based on a discrete direction change sequence.
Background
The intersection point, which is generated when a plurality of curves intersect, is called a curvature discontinuity point. When the curve equation is known, the intersection point can be easily calculated, however, in the practical application of image processing, the curves are usually composed of discrete pixel groups, which brings great difficulty to the detection of curvature discontinuity.
In the field of image processing, the detection of curvature discontinuities plays a crucial role for the segmentation of stacked populations. In practical applications such as astronomical star detection, lesion cell identification, machining and the like, image segmentation is the basis and key, and the accuracy and reliability of results are directly influenced by the quality of segmentation. The traditional image segmentation means that an object is separated from the background by a certain method, however, in practical application in a specific field, the object separated from the background is not single, but a group formed by stacking a plurality of monomers, and the object needs to be segmented into single individuals for further research. The traditional segmentation method can only separate the target from the background, but cannot accurately segment the stacked population, and particularly brings greater difficulty to the segmentation of the stacked population under the complex conditions that some images are easily affected by illumination, edge blurring exists and the like. Therefore, the detection of curvature discontinuity has far-reaching practical significance.
Disclosure of Invention
The invention aims to overcome the defects and shortcomings of the existing curvature discontinuous point detection technology, provides a curvature discontinuous point detection method based on a discrete direction change sequence, can overcome the defects of an image and effectively detect the curvature discontinuous point in a curve formed by discrete pixel points. In the field of image processing, an effective method and an effective thought are provided for the segmentation of a stacked group.
In order to achieve the purpose, the invention adopts the following technical scheme:
a method of curvature discontinuity detection based on a sequence of discrete direction changes, the method comprising the steps of:
s1 preprocessing, searching the end point of the obtained target edge skeleton and the existing Y-shaped connecting points, and removing the Y-shaped connecting points;
s2, obtaining an edge skeleton direction change sequence;
s3 detects the position of the V-shaped connection point.
Specifically, the step S1 is as follows:
s1.1, convolving the acquired skeleton image by using a 3 × 3 template with the value of all 1, selecting a point with the corresponding position value of 2 in the skeleton as an end point of the skeleton, and selecting a point with the corresponding position value of 4 in the skeleton as a Y-shaped connecting point;
s1.2, eliminating Y-shaped connection points, converting all the Y-shaped connection points into V-shaped connection points or directly dividing the V-shaped connection points into 3 discrete parts, and performing area opening operation again to ensure that scattered fragments influencing the operation of the subsequent steps cannot be generated after the connection points are eliminated.
Specifically, the step S2 is as follows:
s2.1, defining a template to represent the direction change of the discrete pixel point, and defining a 3 × 3 template by taking the current pixel point as the center because the next pixel point of the current pixel point is necessarily in the eight-neighborhood, wherein the right value of the center is 1, and the template is increased in a counterclockwise manner until the lower right value is 8;
s2.2, starting from an optional end point to traverse the pixel points in the skeleton, and determining the direction change of the next pixel point to the current pixel point according to a defined direction template so as to obtain the direction change sequences of all the pixel points in the skeleton;
and S2.3, traversing the skeleton direction change sequence obtained in the step S2.2, judging whether the difference between two adjacent numerical values is greater than 4, and if the difference is greater than 4, adding 8 or subtracting 8 to the subsequent numerical value to enable the difference between all the adjacent numerical values to be not greater than 4, so as to obtain a corrected skeleton direction change sequence.
Specifically, the step S3 is as follows:
s3.1, processing the framework direction change sequence, wherein the minimum direction change in the defined direction template is 45 degrees, so that the obtained framework direction change sequence is in a discrete step shape, and proper smoothing processing needs to be carried out on the sequence;
s3.2, taking the average value of the maximum range of the sequence without the V-shaped connection points and the minimum range of the sequence with the V-shaped connection points in the image as a discrimination threshold value to determine whether the V-shaped connection points exist in the framework;
s3.3, determining the position of the V-shaped connecting point, taking the position of the value farthest from the mean value and the position of the value next farthest in the same direction as the position of the V-shaped connecting point, and considering the continuous response within 5 pixels as the same response because the responses of several continuous absolute value maximum values may appear at the V-shaped connecting point.
It should be noted that, the step S3.1 specifically includes:
s3.1.1, local summation processing is carried out on the framework direction change sequence obtained in the step S2.3 to obtain a processed sequence, and the processed sequence is changed into a relatively smooth curve from an original ladder shape;
s3.1.2, differentiating the skeleton direction change sequence obtained in step S3.1.1 to obtain a processed sequence, wherein the processed sequence can reflect the change trend of the previous sequence;
s3.1.3 the skeleton direction change sequence obtained from step S3.1.2 is summed locally to obtain a processed sequence that smoothes the spikes caused by noise while highlighting the spikes caused by the V-junctions.
The method has the advantages that a direction template is defined to express the direction change of the next pixel point to the current pixel point, so that the direction change sequence of the whole framework is obtained, and then proper smoothing and derivation operations are carried out on the sequence, so that the processed sequence can smooth the peak generated by noise and simultaneously highlight the peak generated by a V-shaped connection point, the automatic detection of curvature discontinuous points is realized, and an effective method is provided for the division of stacked groups in the image.
Drawings
FIG. 1 is a schematic diagram of a structure for detecting curvature discontinuities;
FIG. 2 is a schematic diagram of a direction template defined for representing direction changes of discrete pixels;
FIG. 3 is a diagram illustrating a modified sequence of changes in orientation of the skeleton;
FIG. 4 is a schematic diagram illustrating a comparison of smooth derivation and then smoothing of a sequence of changes in direction of a skeleton;
FIG. 5 shows the location of the "V" connection point in the skeleton determined by the present invention for this example.
Detailed Description
The present invention will be further described with reference to the accompanying drawings, and it should be noted that the following examples are provided to illustrate the detailed embodiments and specific operations based on the technical solutions of the present invention, but the scope of the present invention is not limited to the examples.
As shown in fig. 1, the present invention is a method for detecting curvature discontinuity point based on discrete direction change sequence, the method includes the following steps:
s1 preprocessing, searching the end point of the obtained target edge skeleton and the existing Y-shaped connecting points, and removing the Y-shaped connecting points;
s2, obtaining an edge skeleton direction change sequence;
s3 detects the position of the V-shaped connection point.
Specifically, the step S1 is as follows:
s1.1, convolving the acquired skeleton image by using a 3 × 3 template with the value of all 1, selecting a point with the corresponding position value of 2 in the skeleton as an end point of the skeleton, and selecting a point with the corresponding position value of 4 in the skeleton as a Y-shaped connecting point;
s1.2, eliminating Y-shaped connection points, converting all the Y-shaped connection points into V-shaped connection points or directly dividing the V-shaped connection points into 3 discrete parts, and performing area opening operation again to ensure that scattered fragments influencing the operation of the subsequent steps cannot be generated after the connection points are eliminated.
Specifically, the step S2 is as follows:
s2.1, defining a template to represent the direction change of the discrete pixel point, and defining a 3 × 3 template by taking the current pixel point as the center because the next pixel point of the current pixel point is necessarily in the eight-neighborhood, wherein the right value of the center is 1, and the template is increased in a counterclockwise manner until the lower right value is 8;
s2.2, starting from an optional end point to traverse the pixel points in the skeleton, and determining the direction change of the next pixel point to the current pixel point according to a defined direction template so as to obtain the direction change sequences of all the pixel points in the skeleton;
and S2.3, traversing the skeleton direction change sequence obtained in the step S2.2, judging whether the difference between two adjacent numerical values is greater than 4, and if the difference is greater than 4, adding 8 or subtracting 8 to the subsequent numerical value to enable the difference between all the adjacent numerical values to be not greater than 4, so as to obtain a corrected skeleton direction change sequence.
Specifically, the step S3 is as follows:
s3.1, processing the framework direction change sequence, wherein the minimum direction change in the defined direction template is 45 degrees, so that the obtained framework direction change sequence is in a discrete step shape, and proper smoothing processing needs to be carried out on the sequence;
s3.2, taking the average value of the maximum range of the sequence without the V-shaped connection points and the minimum range of the sequence with the V-shaped connection points in the image as a discrimination threshold value to determine whether the V-shaped connection points exist in the framework;
s3.3, determining the position of the V-shaped connecting point, taking the position of the value farthest from the mean value and the position of the value next farthest in the same direction as the position of the V-shaped connecting point, and considering the continuous response within 5 pixels as the same response because the responses of several continuous absolute value maximum values may appear at the V-shaped connecting point.
It should be noted that, the step S3.1 specifically includes:
s3.1.1, local summation processing is carried out on the framework direction change sequence obtained in the step S2.3 to obtain a processed sequence, and the processed sequence is changed into a relatively smooth curve from an original ladder shape;
s3.1.2, differentiating the skeleton direction change sequence obtained in step S3.1.1 to obtain a processed sequence, wherein the processed sequence can reflect the change trend of the previous sequence;
s3.1.3 the skeleton direction change sequence obtained from step S3.1.2 is summed locally to obtain a processed sequence that smoothes the spikes caused by noise while highlighting the spikes caused by the V-junctions.
Examples
Generally speaking, the present invention includes the steps of preprocessing step S1, obtaining the edge skeleton direction change sequence in step S2, and detecting the position of the "V-shaped" connection point in step S3, i.e. as shown in fig. 1.
The preprocessing step comprises 2 sub-steps of searching for the end point and the Y-shaped connecting point of the skeleton, eliminating the Y-shaped connecting point and the like; the step of obtaining the change sequence of the direction of the edge framework comprises 3 sub-steps of defining a direction template, traversing pixel points in the framework to obtain the change direction, correcting the change sequence of the direction of the framework and the like; the step of detecting the position of the V-shaped connecting point comprises 3 sub-steps of processing the direction change sequence of the framework, judging whether the V-shaped connecting point exists or not, and determining the position of the V-shaped connecting point. Specifically, the method comprises the following steps:
step S1, preprocessing
Firstly, convolving the acquired skeleton image by using a 3 × 3 template with the value of all 1, selecting a point with the corresponding position value of 2 in the skeleton as an end point of the skeleton, and selecting a point with the corresponding position value of 4 in the skeleton as a Y-shaped connecting point.
And secondly, eliminating the Y-shaped connecting points to convert all the Y-shaped connecting points into V-shaped connecting points or directly divide the V-shaped connecting points into 3 discrete parts, and finally performing the region opening operation once again to ensure that scattered fragments influencing the operation of the subsequent steps cannot be generated after the connecting points are eliminated.
Step S2, obtaining the edge skeleton direction change sequence
First, a template is defined to represent the direction change of the discrete pixel, and since the next pixel of the current pixel must be in its eight neighbors, a template of 3 × 3 is defined with the pixel as the center, the right value of the center is 1, and the template increases counterclockwise until the lower right value is 8, and the direction template is as shown in fig. 2.
And secondly, starting from an optional end point to traverse the pixel points in the skeleton, and determining the direction change of the next pixel point to the current pixel point according to the defined direction template so as to obtain the direction change sequence of all the pixel points in the skeleton.
Finally, the framework orientation change sequence is corrected. The above direction definition has a drawback that the directions indicated by 1 and 8 are two adjacent directions, but are different in value by 7, and similarly, all the directions crossing the boundary between 1 and 8 and having a manhattan distance of less than 4 in the direction definition lattice points, such as 1 and 7, 2 and 8, cannot represent the difference in direction by the difference in value, which results in that the position of the "V-shaped" connection point cannot be directly judged by the abrupt change in value. It is reasonable to say that the difference in value is proportional to the difference in direction, and every 45 degrees in angle, the value differs by 1. Traversing the skeleton direction change sequence, judging whether the difference between two adjacent numerical values is greater than 4, if so, adding 8 or subtracting 8 to the following numerical values to enable the difference between all the adjacent numerical values to be not greater than 4, and obtaining the corrected skeleton direction change sequence as shown in figure 3. This operation will change the range of the sequence but is advantageous for the determination of "V-shaped" junctions.
Step S3, detecting the position of the V-shaped connection point
First, a sequence of changes in the orientation of the backbone is processed. Since the minimum direction change in the defined direction template is 45 °, the obtained skeleton direction change sequence is in a discrete staircase shape. Because the framework direction change sequence is not in a perfect step shape and has randomness, the local mean processing is performed to eliminate the randomness to a certain extent, the result sequences obtained by the local mean processing and the local summation processing in the application are completely consistent except for a proportionality coefficient, the local summation processing can be used for keeping the minimum interval of numerical values in the sequence to be 1, so that the result is convenient to analyze, the operation amount can be reduced, the operation speed can be improved, and the change condition of an angle can be reflected even though the numerical values on the sequence do not represent the change of the angle any more, so that the local mean is replaced by the local summation. The sequence after the local summation processing is changed into a relatively smooth curve from the original ladder shape. The sequence is subjected to differential processing to acquire the variation trend of the sequence. Assuming that there is a continuous ideal circular arc, the skeleton direction change sequence and the sequence after local summation should be straight line segments, and the straight line segments with constant slope are obtained after differentiation. Thus, the position where the sequence slope is not 0 after differentiation must be a "V-shaped" junction. However, since the image is discrete and has noise influence, in the absence of "V-shaped" connection points, the sequence after differentiation actually obtained is such that most of the values jump (discrete) between-1 and 1, and there is a small probability that a peak (noise) jumping out of the range of-1 to 1 occurs, and the absolute value of the peak is 2. And at the "V-shaped" junction there appears a sharp peak, the absolute value of the most significant of which is 2 or 3. When the absolute value of the peak maximum is 2, the noise and the "V-shaped" connection point cannot be distinguished by the absolute value of the peak maximum. Since the peaks of the "V-shaped" junctions are wider than the peaks caused by noise and are "convex", the sequence is again subjected to a local summation process in order to smooth the peaks caused by noise while highlighting the peaks caused by the "V-shaped" junctions. The image obtained by the three-pass processing is shown in fig. 4.
Second, it is determined whether a "V-shaped" connection point exists in the skeleton. After the last step, there is a sharp peak at the position of the "V-shaped" connection point in the sequence, and the absolute value of the sharp peak is much larger than that of the mean value, which results in that the extreme difference in the sequence is significantly larger than that in the case that there is no "V-shaped" connection point. Through tests, the maximum pole difference of the sequences without the V-shaped cross points in all the test images is smaller than the minimum pole difference of the sequences with the V-shaped cross points, and finally the average value of the maximum pole difference is taken as a discrimination threshold value to determine whether the V-shaped connection points exist in the framework.
Finally, the location of the "V" connection point is determined. After the V-shaped connection points are determined to exist, because the number of the V-shaped connection points in the skeleton is not determined, the positions of the value farthest from the mean value and the value second farthest in the same direction are searched as the positions of the V-shaped connection points, and because continuous responses within 5 pixels can occur at the V-shaped connection points, the responses of the maximum values of several continuous absolute values can be regarded as the same response. The position of the "V" connection point in the skeleton is shown in FIG. 5 (red circle portion).
Through all the steps, all the V-shaped connecting points of the edge framework can be found out. The obtained connection points are utilized to segment the image edges and respectively carry out circle fitting on the edges, so that the final purpose of the embodiment can be realized: false detection of stars in the astronomical images is excluded. If the detected point is located within any of the fitted circles, the point is marked as a false detection.
Various corresponding changes and modifications can be made by those skilled in the art based on the above technical solutions and concepts, and all such changes and modifications should be included in the protection scope of the present invention.
Claims (5)
1. A method for detecting a curvature discontinuity point based on a sequence of discrete direction changes, the method comprising the steps of:
s1 preprocessing, searching the end point of the obtained target edge skeleton and the existing Y-shaped connecting points, and removing the Y-shaped connecting points;
s2, obtaining an edge skeleton direction change sequence;
s3 detects the position of the V-shaped connection point.
2. The method for detecting curvature discontinuity point based on discrete direction change sequence according to claim 1, wherein the step S1 specifically comprises:
s1.1, convolving the acquired skeleton image by using a 3 × 3 template with the value of all 1, selecting a point with the corresponding position value of 2 in the skeleton as an end point of the skeleton, and selecting a point with the corresponding position value of 4 in the skeleton as a Y-shaped connecting point;
s1.2, eliminating Y-shaped connection points, converting all the Y-shaped connection points into V-shaped connection points or directly dividing the V-shaped connection points into 3 discrete parts, and performing area opening operation again to ensure that scattered fragments influencing the operation of the subsequent steps cannot be generated after the connection points are eliminated.
3. The method for detecting curvature discontinuity point based on discrete direction change sequence according to claim 1, wherein the step S2 specifically comprises:
s2.1, defining a template to represent the direction change of the discrete pixel point, and defining a 3 × 3 template by taking the current pixel point as the center because the next pixel point of the current pixel point is necessarily in the eight-neighborhood, wherein the right value of the center is 1, and the template is increased in a counterclockwise manner until the lower right value is 8;
s2.2, starting from an optional end point to traverse the pixel points in the skeleton, and determining the direction change of the next pixel point to the current pixel point according to a defined direction template so as to obtain the direction change sequences of all the pixel points in the skeleton;
and S2.3, traversing the skeleton direction change sequence obtained in the step S2.2, judging whether the difference between two adjacent numerical values is greater than 4, and if the difference is greater than 4, adding 8 or subtracting 8 to the subsequent numerical value to enable the difference between all the adjacent numerical values to be not greater than 4, so as to obtain a corrected skeleton direction change sequence.
4. The method for detecting curvature discontinuity point based on discrete direction change sequence according to claim 1, wherein the step S3 specifically comprises:
s3.1, processing the framework direction change sequence, wherein the minimum direction change in the defined direction template is 45 degrees, so that the obtained framework direction change sequence is in a discrete step shape, and proper smoothing processing needs to be carried out on the sequence;
s3.2, taking the average value of the maximum range of the sequence without the V-shaped connection points and the minimum range of the sequence with the V-shaped connection points in the image as a discrimination threshold value to determine whether the V-shaped connection points exist in the framework;
s3.3, determining the position of the V-shaped connecting point, taking the position of the value farthest from the mean value and the position of the value next farthest in the same direction as the position of the V-shaped connecting point, and considering the continuous response within 5 pixels as the same response because the responses of several continuous absolute value maximum values may appear at the V-shaped connecting point.
5. The method for detecting curvature discontinuity point based on discrete direction change sequence according to claim 4, wherein the step S3.1 specifically comprises:
s3.1.1, local summation processing is carried out on the framework direction change sequence obtained in the step S2.3 to obtain a processed sequence, and the processed sequence is changed into a relatively smooth curve from an original ladder shape;
s3.1.2, differentiating the skeleton direction change sequence obtained in step S3.1.1 to obtain a processed sequence, wherein the processed sequence can reflect the change trend of the previous sequence;
s3.1.3 the skeleton direction change sequence obtained from step S3.1.2 is summed locally to obtain a processed sequence that smoothes the spikes caused by noise while highlighting the spikes caused by the V-junctions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010077829.7A CN111325789B (en) | 2020-02-01 | 2020-02-01 | Curvature discontinuous point detection method based on discrete direction change sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010077829.7A CN111325789B (en) | 2020-02-01 | 2020-02-01 | Curvature discontinuous point detection method based on discrete direction change sequence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111325789A true CN111325789A (en) | 2020-06-23 |
| CN111325789B CN111325789B (en) | 2024-01-09 |
Family
ID=71165185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010077829.7A Active CN111325789B (en) | 2020-02-01 | 2020-02-01 | Curvature discontinuous point detection method based on discrete direction change sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111325789B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115575857A (en) * | 2022-12-08 | 2023-01-06 | 江西广凯新能源股份有限公司 | Emergency protection method and device for high-voltage wire breakage |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0683952A (en) * | 1992-09-01 | 1994-03-25 | Kazuo Toraichi | Device and method for inputting and outputting character data |
| CN103942552A (en) * | 2014-02-11 | 2014-07-23 | 北京大学 | Character image vectorization method and system based on framework instruction |
| CN104697476A (en) * | 2015-03-19 | 2015-06-10 | 北京时代之峰科技有限公司 | Automatic detection method and device for roughness light-incision profile curve |
-
2020
- 2020-02-01 CN CN202010077829.7A patent/CN111325789B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0683952A (en) * | 1992-09-01 | 1994-03-25 | Kazuo Toraichi | Device and method for inputting and outputting character data |
| CN103942552A (en) * | 2014-02-11 | 2014-07-23 | 北京大学 | Character image vectorization method and system based on framework instruction |
| CN104697476A (en) * | 2015-03-19 | 2015-06-10 | 北京时代之峰科技有限公司 | Automatic detection method and device for roughness light-incision profile curve |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115575857A (en) * | 2022-12-08 | 2023-01-06 | 江西广凯新能源股份有限公司 | Emergency protection method and device for high-voltage wire breakage |
| CN115575857B (en) * | 2022-12-08 | 2023-04-28 | 江西广凯新能源股份有限公司 | Emergency protection method and device for high-voltage wire breakage |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111325789B (en) | 2024-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114140462B (en) | Bearing wear degree assessment method based on image processing | |
| EP0149457B1 (en) | Method of identifying contour lines | |
| CN107679520B (en) | Lane line visual detection method suitable for complex conditions | |
| JP6487576B2 (en) | Fingerprint ridge point recognition method and apparatus | |
| CN115511889A (en) | Method for detecting welding defects on surface of solar cell panel bracket | |
| CN116091504B (en) | Connecting pipe connector quality detection method based on image processing | |
| JPH0351968A (en) | Linearization decision system | |
| US5923776A (en) | Object extraction in images | |
| CN110659645B (en) | Character recognition method for digital instrument | |
| CN111667470B (en) | Industrial pipeline flaw detection inner wall detection method based on digital image | |
| CN115311275B (en) | Method and system for detecting surface defects of electronic component | |
| CN116721107B (en) | Intelligent monitoring system for cable production quality | |
| CN112508913A (en) | Cable section edge detection method based on image detection | |
| CN112101351A (en) | Projection-based text line rotation correction method and device | |
| CN115063430A (en) | Electric pipeline crack detection method based on image processing | |
| CN114863492A (en) | Method and device for repairing low-quality fingerprint image | |
| CN111325789A (en) | Curvature discontinuity point detection method based on discrete direction change sequence | |
| CN116188468A (en) | HDMI cable transmission letter sorting intelligent control system | |
| CN113686874B (en) | Mechanical part damage detection method and system based on artificial intelligence | |
| CN1218279C (en) | Segmentation of digital images | |
| CN112270679B (en) | Image segmentation method for convex polygon outline combining concave points and concave edges | |
| CN111583178A (en) | Method for segmenting adhesive algae cells in algae cell microscopic image | |
| Nasr-Isfahani et al. | A new approach for touching cells segmentation | |
| US20240020845A1 (en) | Image Shadow Detection Method and System, and Image Segmentation Device and Readable Storage Medium | |
| CN113221898B (en) | Automatic water level gauge reading method |
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 |