CN117969535A - Road bridge concrete defect nondestructive testing method and system - Google Patents
Road bridge concrete defect nondestructive testing method and system Download PDFInfo
- Publication number
- CN117969535A CN117969535A CN202410384699.XA CN202410384699A CN117969535A CN 117969535 A CN117969535 A CN 117969535A CN 202410384699 A CN202410384699 A CN 202410384699A CN 117969535 A CN117969535 A CN 117969535A
- Authority
- CN
- China
- Prior art keywords
- pixel
- sequence
- road
- bridge concrete
- pixel sequence
- 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
- 238000000034 method Methods 0.000 title claims abstract description 64
- 230000007547 defect Effects 0.000 title claims abstract description 52
- 238000009659 non-destructive testing Methods 0.000 title claims abstract description 19
- 238000011156 evaluation Methods 0.000 claims abstract description 29
- 238000007789 sealing Methods 0.000 claims abstract description 29
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 16
- 230000009467 reduction Effects 0.000 claims description 19
- 238000001228 spectrum Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 10
- 230000003044 adaptive effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000006378 damage Effects 0.000 claims description 5
- 230000001066 destructive effect Effects 0.000 claims 1
- 238000007689 inspection Methods 0.000 claims 1
- 238000010183 spectrum analysis Methods 0.000 abstract description 4
- 241000264877 Hippospongia communis Species 0.000 description 32
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 14
- 238000004422 calculation algorithm Methods 0.000 description 12
- 238000004364 calculation method Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000003763 carbonization Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000009746 freeze damage Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 101150065184 sym-2 gene Proteins 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/766—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using regression, e.g. by projecting features on hyperplanes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Analytical Chemistry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Chemical & Material Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Software Systems (AREA)
- Remote Sensing (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Signal Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to the technical field of spectrum analysis, and provides a road and bridge concrete defect nondestructive testing method and system, wherein the method comprises the following steps: obtaining road and bridge concrete hyperspectral data; acquiring a pixel cutting area sealing coefficient and an intensity deviation coefficient of a pixel sequence according to road bridge concrete hyperspectral data, and further acquiring a pixel honeycomb intensity evaluation coefficient of the pixel sequence; acquiring a pixel waveform disturbance homogeneity evaluation coefficient of a pixel sequence, acquiring a pixel waveform distortion disturbed index of the pixel sequence, and further acquiring the self-adaptive wavelet decomposition layer number of the pixel sequence; and acquiring the concrete integrity of the road bridge concrete, judging the quality of the road bridge concrete according to the concrete integrity, and feeding back to technicians to finish nondestructive detection of the road bridge concrete defects. The method solves the problem that hyperspectral data are over-fitted due to different degrees of influence of noise on pixel sequences, so that road and bridge concrete defect detection is inaccurate.
Description
Technical Field
The invention relates to the technical field of spectrum analysis, in particular to a road and bridge concrete defect nondestructive testing method and system.
Background
In the process of road and bridge concrete pouring and using, defects such as cracks, honeycombs, hollows, freeze injury, concrete carbonization, alkali-aggregate reaction and the like may exist, the structural strength and the service life of the road and bridge concrete are seriously influenced, and therefore the defects of the road and bridge concrete need to be detected after the road and bridge concrete is poured. The defect detection of concrete is usually realized by adopting spectrum analysis, but in the process of detecting the defect of the concrete by adopting spectrum analysis, the number of wavelengths of a spectrometer is more, and the spectrometer is easily influenced by environmental noise and the defect of the concrete, so that a larger gap exists between a pixel sequence obtained by defect detection and an actual pixel. Therefore, noise reduction treatment is required to be carried out on the road and bridge concrete hyperspectral data acquired by the spectrometer.
The wavelet threshold denoising method is simple in algorithm, small in calculated amount and convenient to realize, and noise reduction of road and bridge concrete hyperspectral data is realized, however, the number of decomposition layers of a sequence in the wavelet threshold denoising is a fixed set value, noise suppression cannot be carried out according to the degree of influence of noise on a pixel sequence, and the defect of over fitting or artifact is easily caused.
Disclosure of Invention
The invention provides a road and bridge concrete defect nondestructive testing method and a system, which aim to solve the problem that hyperspectral data are over-fitted due to different degrees of influence of noise on pixel sequences, so that road and bridge concrete defect detection is inaccurate, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides a method for nondestructive testing of road and bridge concrete defects, the method comprising the steps of:
Acquiring road and bridge concrete hyperspectral data and deleting data of a background area;
Acquiring a pixel spectrum curve according to road bridge concrete hyperspectral data, acquiring tangent lines of the pixel spectrum curve on each wavelength, respectively acquiring arc cutting areas of each wavelength, further acquiring pixel arc cutting area sealing coefficients of pixel sequences, determining a comparison pixel set of pixels, acquiring an intensity deviation coefficient according to the road bridge concrete hyperspectral data, and acquiring a pixel honeycomb intensity evaluation coefficient of the pixel sequences according to the intensity deviation coefficient, the pixel arc cutting area sealing coefficients, the pixel sequences and the comparison pixel set of the pixels;
Acquiring pixel waveform disorder homogeneity assessment coefficients of a pixel sequence according to a comparison pixel set of the pixel, a pixel cutting area sealing coefficient of the pixel sequence and an intensity deviation coefficient among the pixel sequence, acquiring pixel waveform distortion disturbance indexes of the pixel sequence according to reflection intensities corresponding to all wavelengths of the pixel sequence, the pixel waveform disorder homogeneity assessment coefficients of the pixel sequence and a pixel honeycomb intensity assessment coefficient, and further acquiring the adaptive wavelet decomposition layer number of the pixel sequence;
Acquiring the concrete integrity of the road bridge concrete according to the pixel sequence after noise reduction and the pixel sequence in the hyperspectral data of the road bridge concrete with equal length and no defect damage, judging the quality of the road bridge concrete according to the concrete integrity, and feeding back to technicians to finish nondestructive detection of the road bridge concrete defect;
the specific method for acquiring the pixel cutting area sealing coefficient of the pixel sequence comprises the following steps:
The sum of arc cutting areas of the pixel spectrum curves of the pixel sequence at all wavelengths is recorded as a pixel arc cutting area sealing coefficient of the pixel sequence;
the specific method for acquiring the intensity deviation coefficient comprises the following steps:
The method comprises the steps of marking a pixel sequence in a comparison pixel set of pixels as a comparison pixel sequence of the pixels, marking the absolute value of the difference value of the reflection intensity of the pixel and the comparison pixel sequence at the same wavelength as the reflection intensity difference of the pixel at the wavelength, marking the sum of the reflection intensity differences of the pixels at all wavelengths as the intensity deviation coefficient of the pixel and the comparison pixel sequence;
the specific method for acquiring the pixel waveform disorder homogeneity evaluation coefficient of the pixel sequence comprises the following steps:
in the method, in the process of the invention, A pixel waveform disorder homogeneity assessment coefficient representing an ith pixel sequence; /(I)Representing the number of pixel sequences contained in the ith pixel comparison pixel set; /(I)Representing the number of rows of two pixel sequences randomly selected from the comparison pixel set of the ith pixel; /(I)In the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the mth pixel sequence; /(I)In the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the nth pixel sequence is shown; /(I)In the pixel comparison set of the ith pixel sequence, the mth pixel sequence/>And nth pixel sequence/>Is the euclidean distance of (2); /(I)An exponential function based on a natural number; /(I)Representing the intensity deviation coefficient of the mth pixel sequence and the nth pixel sequence in the comparison pixel set of the ith pixel;
the specific method for acquiring the pixel honeycomb intensity evaluation coefficient comprises the following steps:
in the method, in the process of the invention, A pixel cell intensity evaluation coefficient representing an ith pixel sequence in the hyperspectral data; /(I)Representing the number of pixel sequences contained in the ith pixel comparison pixel set; /(I)Representing the intensity deviation coefficient of the ith pixel in the hyperspectral data and the mth pixel sequence in the pixel comparison set; /(I)Representing the i-th sequence of picture elements/>The mth pixel sequence/>, in the set, is compared with the pixelsDTW distance between; /(I)An exponential function based on a natural number e; /(I)Representing a pixel cutting area sealing coefficient of an ith pixel sequence in hyperspectral data; /(I)And in the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the mth pixel sequence is shown.
Further, the method for acquiring the contrast pixel set of the pixels comprises the following steps:
taking the pixel sequence of the preset number of pixels nearest to the pixel as a comparison pixel set of the pixels.
Further, the method for acquiring the pixel waveform distortion disturbed index of the pixel sequence comprises the following steps:
recording the sum of information entropy of the pixel waveform disorder homogeneity assessment coefficient of the pixel sequence and the reflection intensity corresponding to all wavelengths of the pixel sequence as a first sum value of the pixel sequence;
Taking the natural number as a base number, taking the pixel honeycomb intensity evaluation coefficient of the pixel sequence as the power of an exponent, and recording the power as a first power value of the pixel sequence;
And recording the product of the first sum value and the first power value of the pixel sequence as a pixel waveform distortion disturbed index of the pixel sequence.
Further, the method for obtaining the adaptive wavelet decomposition layer number of the pixel sequence comprises the following specific steps:
And marking the rounded value of the pixel waveform distortion disturbed index of the pixel sequence and the minimum value in the first parameter adjusting factor as the self-adaptive wavelet decomposition layer number of the pixel sequence.
Further, the concrete integrity of the road and bridge concrete is obtained according to the pixel sequence after noise reduction and the pixel sequence in the hyperspectral data of the road and bridge concrete with equal length and no defect, and the concrete integrity comprises the following specific methods:
Cosine similarity of pixel sequences in hyperspectral data of road and bridge concrete after noise reduction and pixel sequences in hyperspectral data of road and bridge concrete without defect of equal length is recorded as complete similarity of pixel sequences after noise reduction;
and (3) marking the average value of the complete similarity of all the pixel sequences after noise reduction in all the sampling positions in the hyperspectral data of the road bridge concrete as the concrete integrity of the road bridge concrete.
Further, the quality of the bridge concrete is judged according to the integrity of the concrete and is fed back to technicians, and nondestructive testing of the bridge concrete defect is completed, and the concrete method comprises the following steps:
When the concrete integrity of the road and bridge concrete is greater than or equal to an integrity threshold value, the quality of the road and bridge concrete is considered to be qualified, otherwise, the quality of the road and bridge concrete is considered to be unqualified;
And feeding back the quality judgment result of the road and bridge concrete to engineering acceptance technicians, and realizing nondestructive detection of the road and bridge concrete defects.
In a second aspect, an embodiment of the present invention further provides a road and bridge concrete defect nondestructive testing system, including a memory, a processor, and a computer program stored in the memory and running on the processor, where the processor implements the steps of any one of the methods described above when executing the computer program.
The beneficial effects of the invention are as follows:
According to the method, the problem that concrete is locally loosened due to uneven concrete pouring or insufficient vibration of roads and bridges, concrete is easy to generate concrete honeycombs is solved, according to the structural characteristics that each wavelength of a pixel spectrum curve has larger random fluctuation, the degree of influence of noise on a pixel sequence is estimated, the intensity deviation coefficient is obtained, then, according to the characteristics that honeycombs generated in the road and bridge concrete are scattered, other pixels around pixels corresponding to a honeycomb area correspond to a normal concrete area, analysis is continued, and the pixel honeycomb intensity evaluation coefficient of a pixel sequence is obtained; further, considering that the environmental noise and the honeycomb phenomenon exist simultaneously, when the pixel is a normal pixel with smaller noise interference degree, but other pixels around the pixel correspond to the position of the honeycomb defect or have larger noise interference degree, the normal pixel is easy to be misjudged as the pixel with the defect to cause misjudgment, the pixel waveform disorder homogeneity evaluation coefficient of the pixel sequence is obtained, and the misjudgment condition is accurately identified; according to the noise interference degree of the pixels, the self-adaptive wavelet decomposition layer number of the pixel sequence is obtained, the smaller self-adaptive wavelet decomposition layer number is selected for the pixels with smaller noise interference degree, the calculation speed of wavelet threshold denoising is improved, the larger self-adaptive wavelet decomposition layer number is selected for the pixels with larger noise interference degree, the pixel sequence is decomposed more accurately, more detail information is reserved, and the noise reduction accuracy degree is improved; and finally, denoising the pixel sequence according to the self-adaptive wavelet decomposition layer number of the pixel sequence, judging the quality of the bridge concrete, finishing the nondestructive detection of the bridge concrete defect, and solving the problem that the bridge concrete defect detection is inaccurate due to the fact that hyperspectral data are over-fitted due to different degrees of influence of noise on the pixel sequence.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for nondestructive testing of road and bridge concrete defects according to an embodiment of the present invention;
Fig. 2 is a flowchart of the intensity deviation coefficient acquisition.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flowchart of a method for nondestructive testing of road and bridge concrete defects according to an embodiment of the invention is shown, the method includes the following steps:
and S001, acquiring road and bridge concrete hyperspectral data and deleting data of a background area.
After the road and bridge concrete is poured, carrying a full-band hyperspectral imager HG-HyperUAV-2500 on the fixed wing unmanned aerial vehicle, carrying out hyperspectral detection on the poured road and bridge, and sampling at equal intervals to obtain the hyperspectral data of the road and bridge concrete.
In order to ensure the accuracy of the sampled hyperspectral data, the unmanned aerial vehicle is hovered at a position 5m above the road bridge horizontal plane, and the sampling interval is set to be 1 meter.
According to the road and bridge concrete hyperspectral data, the road and bridge concrete hyperspectral data of each sampling position can be obtained, the road and bridge concrete hyperspectral data comprise a plurality of pixel sequences, and each pixel sequence has corresponding reflection intensity information at each wavelength.
When the unmanned aerial vehicle is suspended to scan data, the background information outside the road and bridge concrete boundary can be scanned due to wider visual angles, and interference of the background information on concrete defect detection needs to be eliminated.
And respectively taking the hyperspectral data of each single sampling position as the input of GRBS true color synthesis algorithm, and outputting the algorithm as a pseudo-color image corresponding to the current sampling moment. In a pseudo-color image, the pixel position information remains unchanged. And taking the pseudo-color image corresponding to the current sampling moment as input of a Mask2Former semantic segmentation neural network model, extracting a classification region of the pseudo-color image through the semantic segmentation neural network model, and dividing pixels in road bridge concrete hyperspectral data into a background region and a concrete region. GRBS true color synthesis algorithm and Mask2Former semantic segmentation neural network model are known technologies and will not be described in detail.
And deleting the road and bridge concrete hyperspectral data of the background area, and only analyzing the road and bridge concrete hyperspectral data in the concrete area to eliminate the interference of the background area.
So far, the hyperspectral data of the road and bridge concrete are obtained.
Step S002, a comparison pixel set of pixels is determined, a pixel cutting area sealing coefficient and an intensity deviation coefficient of a pixel sequence are obtained according to road bridge concrete hyperspectral data, and a pixel honeycomb intensity evaluation coefficient of the pixel sequence is obtained.
In the process of pouring road and bridge concrete, the phenomenon of concrete honeycomb can occur due to partial loosening of the interior caused by uneven pouring or insufficient vibration. The internal structure of the concrete honeycomb is complex, the hyperspectral data sampling is easy to generate random influence, and the spectrum detection precision is reduced.
In the wavelet threshold denoising algorithm, the number of decomposition layers of the wavelet determines the degree of detail of the signal being decomposed, but the conventional wavelet threshold denoising algorithm adopts a fixed number of decomposition layers. The larger number of decomposition layers can provide finer signal representation, which is helpful for more accurately identifying and suppressing noise, but the larger number of decomposition layers can cause increased calculation amount, and signal overfitting is easy to cause data loss after denoising to have excessive detail characteristics, and meanwhile, the excessive number of decomposition layers easily causes signal artifacts in wavelet reconstruction, so that the denoising effect is poor. Therefore, the self-adaptive adjustment of the decomposition layer number is required to be carried out by combining the influence degree of noise on the pixel sequence, so that the optimal noise reduction of the pixel sequence is realized.
In order to better analyze the interference condition of noise and honeycomb phenomenon on a single pixel sequence, each pixel sequence is used as the input of a local weighted regression LWR algorithm, and the output of the algorithm is a pixel spectrum curve obtained by fitting the pixel sequences. Sequence the ith pixelObtained/>, for the pixel spectrum curveAnd (3) representing. The LWR algorithm is a known technique, and the embodiment is not described in detail.
Tangential lines are respectively made at the positions of the points corresponding to each wavelength on the pixel spectrum curve, and the tangential lines of the pixel spectrum curve on each wavelength are obtained. Because the spectrum curve of the pixel is an arc line, the tangent lines of two adjacent wavelengths have to have an intersection point, and at the same time, the firstWavelength and/>The abscissa of intersection of the individual wavelengths is at/>Within the range.
And respectively taking the pixel spectrum curve at each wavelength as a wavelength to be analyzed, marking the next adjacent wavelength of the wavelength to be analyzed as the adjacent wavelength of the wavelength to be analyzed, and marking a tangent broken line formed by tangents of the adjacent wavelengths of the wavelength to be analyzed as a tangent broken line of the wavelength to be analyzed. And respectively taking the wavelength to be analyzed and the adjacent wavelength of the wavelength to be analyzed as an integral lower limit and an integral upper limit, integrating the area between the pixel spectrum curve and the tangent fold line of the wavelength to be analyzed, and recording the absolute value of the integral value as the arc cutting area of the wavelength to be analyzed. And (3) recording the sum of the arc cutting areas of all wavelengths to be analyzed of the pixel sequence as a pixel arc cutting area sealing coefficient of the pixel sequence.
When each wavelength of the pixel spectrum curve is seriously affected by noise, larger random fluctuation exists, so that the pixel spectrum curve has more fluctuation, and further the tangential area of each wavelength is increased, so that the closed coefficient of the tangential area of the pixels of the pixel sequence is larger.
When the pixels are positioned in the honeycomb area, the influence of the uncertain structure in the honeycomb area can also cause the waveform of the pixel sequence to generate larger fluctuation, and the fluctuation information reflects the structural condition and detail information of the honeycomb in the concrete, so that the information needs to be reserved. Thus, it is necessary to further evaluate the extent to which the sequence of picture elements is affected by noise in combination with the position information of the picture elements.
The image element sequence of 25 image elements nearest to the image element is taken as a comparison image element set of the image elements.
And obtaining the intensity deviation coefficient according to the hyperspectral data of the road and bridge concrete.
In the method, in the process of the invention,Representing the intensity deviation coefficient of the ith pixel in the hyperspectral data and the mth pixel sequence in the pixel comparison set; /(I)Representing the/>, in the ith sequence of picture elements in the hyperspectral dataReflection intensity at each wavelength; /(I)Representing the/>, in the mth pixel sequence, of the comparison set of pixels of the ith pixel sequence in the hyperspectral dataReflection intensity at each wavelength; l represents the number of wavelengths acquired by the full-band hyperspectral imager.
The hovering height of the unmanned aerial vehicle is higher, and a honeycomb area caused by uneven pouring in road and bridge concrete often corresponds to pixels of individual or smaller areas in hyperspectral data. Meanwhile, because the honeycomb in the road and bridge concrete is scattered, other pixels around the pixels corresponding to the honeycomb area should correspond to the normal concrete area. Therefore, the intensity deviation coefficient of the pixel corresponding to the honeycomb problem is large.
The intensity deviation coefficient acquisition flowchart is shown in fig. 2.
And acquiring a pixel honeycomb strength evaluation coefficient of the pixel sequence according to the strength deviation coefficient, the pixel cutting area sealing coefficient, the pixel sequence and the comparison pixel set of the pixel.
In the method, in the process of the invention,A pixel cell intensity evaluation coefficient representing an ith pixel sequence in the hyperspectral data; /(I)Representing the number of pixel sequences contained in the ith pixel comparison pixel set; /(I)Representing the intensity deviation coefficient of the ith pixel in the hyperspectral data and the mth pixel sequence in the pixel comparison set; /(I)Representing the i-th sequence of picture elements/>The mth pixel sequence/>, in the set, is compared with the pixelsDTW distance between; /(I)An exponential function based on a natural number e; /(I)Representing a pixel cutting area sealing coefficient of an ith pixel sequence in hyperspectral data; /(I)And in the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the mth pixel sequence is shown.
The pixel sequence corresponding to the honeycomb problem has larger difference from the pixel sequence contained in the comparison pixel set of the pixels, and the intensity deviation coefficient of the pixels corresponding to the honeycomb problem is larger, so that when the pixels correspond to the positions where the honeycomb problem occurs, the pixel honeycomb intensity evaluation coefficient of the pixel sequence is larger.
So far, the pixel honeycomb intensity evaluation coefficient of the pixel sequence is acquired.
Step S003, acquiring a pixel waveform disturbance homogeneity evaluation coefficient of a pixel sequence, acquiring a pixel waveform distortion disturbed index of the pixel sequence, and further acquiring the adaptive wavelet decomposition layer number of the pixel sequence.
During actual acquisition, ambient noise and cellular phenomena are present at the same time. When the pixels are normal pixels with smaller noise interference degree and the pixels corresponding to the comparison pixel sets of the pixels correspond to the positions of the honeycomb defects or have larger noise interference degree, the honeycomb strength evaluation coefficients of the pixels corresponding to the normal pixels can not accurately reflect the real scene problem, and the misjudgment condition needs to be eliminated.
And acquiring the pixel waveform disorder homogeneity evaluation coefficient of the pixel sequence according to the comparison pixel set of the pixel, the pixel cutting area sealing coefficient of the pixel sequence and the intensity deviation coefficient among the pixel sequences.
In the method, in the process of the invention,A pixel waveform disorder homogeneity assessment coefficient representing an ith pixel sequence; /(I)Representing the number of pixel sequences contained in the ith pixel comparison pixel set; /(I)Representing the number of rows of two pixel sequences randomly selected from the comparison pixel set of the ith pixel; /(I)In the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the mth pixel sequence; /(I)In the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the nth pixel sequence is shown; /(I)In the pixel comparison set of the ith pixel sequence, the mth pixel sequence/>And nth pixel sequence/>Is the euclidean distance of (2); /(I)An exponential function based on a natural number; /(I)Representing the intensity deviation coefficient of the ith pixel in the comparison pixel set, the mth pixel sequence and the nth pixel sequence.
When the pixels are normal pixels with smaller noise interference degree and the pixels corresponding to the comparison pixel sets of the pixels correspond to the positions of the honeycomb defects or have larger noise interference degree, the difference between the pixels corresponding to the comparison pixel sets of the pixels is larger, the difference of the pixel cutting area sealing coefficients between the pixels corresponding to the comparison pixel sets of the pixels is larger, and the intensity deviation coefficient of the pixels corresponding to the comparison pixel sets of the pixels is larger, so that the pixel waveform disorder homogeneity evaluation coefficient of the normal pixels with smaller noise interference degree is larger. When the pixels corresponding to the pixel and the contrast pixel set of the pixels are normal pixels with smaller noise interference degree, the difference between the pixels corresponding to the contrast pixel set of the pixels is smaller, the difference of the sealing coefficients of the pixel cutting areas between the pixels corresponding to the contrast pixel set of the pixels is smaller, and the strength deviation coefficient of the pixels corresponding to the contrast pixel set of the pixels is smaller, so that the waveform disorder homogeneity evaluation coefficient of the pixels of the normal pixels with smaller noise interference degree is smaller.
Therefore, the misjudgment condition can be accurately identified through the pixel waveform disorder homogeneity evaluation coefficient of the pixel sequence.
And acquiring a pixel waveform distortion disturbed index of the pixel sequence according to the reflection intensity corresponding to all wavelengths of the pixel sequence, the pixel waveform disorder homogeneity evaluation coefficient and the pixel honeycomb intensity evaluation coefficient of the pixel sequence, and further acquiring the self-adaptive wavelet decomposition layer number of the pixel sequence.
In the method, in the process of the invention,Representing the number of adaptive wavelet decomposition layers of an ith pixel sequence in hyperspectral data; /(I)A pixel cell intensity evaluation coefficient representing an i-th sequence of pixels; /(I)A pixel waveform disorder homogeneity assessment coefficient representing an ith pixel sequence; information entropy representing reflection intensity corresponding to all wavelengths of the ith pixel sequence; /(I) An exponential function based on a natural number; /(I)Representing a minimum function, acting as the minimum of the comma-separated values in brackets; /(I)A pixel waveform distortion disturbance index representing an ith pixel sequence; /(I)Representing a first parameter adjustment factor, the value of this embodiment is 3; /(I)Representing a rounding function, acting as a rounding value to take the values within the function.
When the pixel is the pixel with smaller noise interference degree, the pixel waveform disorder homogeneity evaluation coefficient of the pixel sequence is smaller, the adaptive wavelet decomposition layer number of the pixel sequence is smaller, at the moment, the smaller wavelet decomposition layer number is selected, the calculation speed of wavelet threshold denoising is improved, otherwise, the larger wavelet decomposition layer number is selected, the pixel sequence is decomposed more accurately, more detail information is reserved, and the accuracy degree of noise reduction is improved.
So far, the number of self-adaptive wavelet decomposition layers of the pixel sequence is acquired.
And S004, obtaining the concrete integrity of the road and bridge concrete, judging the quality of the road and bridge concrete according to the concrete integrity, and finishing the nondestructive detection of the road and bridge concrete defects.
Taking the self-adaptive wavelet decomposition layer number of the pixel sequence as the value of the wavelet decomposition layer number of the wavelet threshold denoising algorithm, taking the pixel sequence as the input of the wavelet threshold denoising algorithm, taking the wavelet base as sym2, and outputting the pixel sequence after noise reduction. The wavelet threshold denoising algorithm is a known technology and will not be described in detail.
And obtaining the concrete integrity of the road and bridge concrete according to the pixel sequence after noise reduction and the pixel sequence in the hyperspectral data of the road and bridge concrete with equal length and no defect.
In the method, in the process of the invention,Representing the concrete integrity of road and bridge concrete; /(I)The pixel sequence after the i noise reduction of the kth sampling position in the hyperspectral data of the road bridge concrete is represented; /(I)Pixel sequence in hyperspectral data representing road and bridge concrete without defect damage, and length and/>The same; /(I)The cosine similarity of two sequences separated by a comma in brackets; /(I)The number of pixel sequences at the kth sampling position in hyperspectral data of road and bridge concrete is represented; /(I)The number of different sampling positions in the hyperspectral data of the road and bridge concrete is represented.
If the construction of the road and bridge concrete meets the design standard of the road and bridge concrete, the pixel sequence after noise reduction has smaller difference with the pixel sequence in the hyperspectral data of the road and bridge concrete without defects and damage, and at the moment, the obtained concrete integrity of the road and bridge concrete is larger.
And when the concrete integrity of the road and bridge concrete is greater than or equal to an integrity threshold value, the quality of the road and bridge concrete is considered to be qualified, otherwise, the quality of the road and bridge concrete is considered to be unqualified. The value of the integrity threshold in this embodiment is 0.8.
And feeding back the quality judgment result of the road and bridge concrete to engineering acceptance technicians, and providing data reference for the technicians.
Thus, the nondestructive detection of the road and bridge concrete defects is realized.
Based on the same inventive concept as the method, the embodiment of the invention also provides a road and bridge concrete defect nondestructive testing system, which comprises a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor realizes the steps of any one of the road and bridge concrete defect nondestructive testing methods when executing the computer program.
The foregoing description of the preferred embodiments of the present invention is not intended to be limiting, but rather, any modifications, equivalents, improvements, etc. that fall within the principles of the present invention are intended to be included within the scope of the present invention.
Claims (7)
1. The nondestructive testing method for the road and bridge concrete defects is characterized by comprising the following steps of:
Acquiring road and bridge concrete hyperspectral data and deleting data of a background area;
Acquiring a pixel spectrum curve according to road bridge concrete hyperspectral data, acquiring tangent lines of the pixel spectrum curve on each wavelength, respectively acquiring arc cutting areas of each wavelength, further acquiring pixel arc cutting area sealing coefficients of pixel sequences, determining a comparison pixel set of pixels, acquiring an intensity deviation coefficient according to the road bridge concrete hyperspectral data, and acquiring a pixel honeycomb intensity evaluation coefficient of the pixel sequences according to the intensity deviation coefficient, the pixel arc cutting area sealing coefficients, the pixel sequences and the comparison pixel set of the pixels;
Acquiring pixel waveform disorder homogeneity assessment coefficients of a pixel sequence according to a comparison pixel set of the pixel, a pixel cutting area sealing coefficient of the pixel sequence and an intensity deviation coefficient among the pixel sequence, acquiring pixel waveform distortion disturbance indexes of the pixel sequence according to reflection intensities corresponding to all wavelengths of the pixel sequence, the pixel waveform disorder homogeneity assessment coefficients of the pixel sequence and a pixel honeycomb intensity assessment coefficient, and further acquiring the adaptive wavelet decomposition layer number of the pixel sequence;
Acquiring the concrete integrity of the road bridge concrete according to the pixel sequence after noise reduction and the pixel sequence in the hyperspectral data of the road bridge concrete with equal length and no defect damage, judging the quality of the road bridge concrete according to the concrete integrity, and feeding back to technicians to finish nondestructive detection of the road bridge concrete defect;
the specific method for acquiring the pixel cutting area sealing coefficient of the pixel sequence comprises the following steps:
The sum of arc cutting areas of the pixel spectrum curves of the pixel sequence at all wavelengths is recorded as a pixel arc cutting area sealing coefficient of the pixel sequence;
the specific method for acquiring the intensity deviation coefficient comprises the following steps:
The method comprises the steps of marking a pixel sequence in a comparison pixel set of pixels as a comparison pixel sequence of the pixels, marking the absolute value of the difference value of the reflection intensity of the pixel and the comparison pixel sequence at the same wavelength as the reflection intensity difference of the pixel at the wavelength, marking the sum of the reflection intensity differences of the pixels at all wavelengths as the intensity deviation coefficient of the pixel and the comparison pixel sequence;
the specific method for acquiring the pixel waveform disorder homogeneity evaluation coefficient of the pixel sequence comprises the following steps:
in the method, in the process of the invention, A pixel waveform disorder homogeneity assessment coefficient representing an ith pixel sequence; /(I)Representing the number of pixel sequences contained in the ith pixel comparison pixel set; /(I)Representing the number of rows of two pixel sequences randomly selected from the comparison pixel set of the ith pixel; /(I)In the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the mth pixel sequence; /(I)In the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the nth pixel sequence is shown; /(I)In the pixel comparison set of the ith pixel sequence, the mth pixel sequence/>And nth pixel sequence/>Is the euclidean distance of (2); /(I)An exponential function based on a natural number; /(I)Representing the intensity deviation coefficient of the mth pixel sequence and the nth pixel sequence in the comparison pixel set of the ith pixel;
the specific method for acquiring the pixel honeycomb intensity evaluation coefficient comprises the following steps:
in the method, in the process of the invention, A pixel cell intensity evaluation coefficient representing an ith pixel sequence in the hyperspectral data; /(I)Representing the number of pixel sequences contained in the ith pixel comparison pixel set; /(I)Representing the intensity deviation coefficient of the ith pixel in the hyperspectral data and the mth pixel sequence in the pixel comparison set; /(I)Representing the i-th sequence of picture elements/>The mth pixel sequence/>, in the set, is compared with the pixelsDTW distance between; /(I)An exponential function based on a natural number e; /(I)Representing a pixel cutting area sealing coefficient of an ith pixel sequence in hyperspectral data; /(I)And in the pixel comparison set of the ith pixel sequence in the hyperspectral data, the pixel cutting area sealing coefficient of the mth pixel sequence is shown.
2. The road and bridge concrete defect nondestructive testing method according to claim 1, wherein the method for obtaining the contrast pixel set of the pixels is as follows:
taking the pixel sequence of the preset number of pixels nearest to the pixel as a comparison pixel set of the pixels.
3. The method for nondestructive testing of road and bridge concrete defects according to claim 1, wherein the method for obtaining the pixel waveform distortion disturbed index of the pixel sequence comprises the following steps:
recording the sum of information entropy of the pixel waveform disorder homogeneity assessment coefficient of the pixel sequence and the reflection intensity corresponding to all wavelengths of the pixel sequence as a first sum value of the pixel sequence;
Taking the natural number as a base number, taking the pixel honeycomb intensity evaluation coefficient of the pixel sequence as the power of an exponent, and recording the power as a first power value of the pixel sequence;
And recording the product of the first sum value and the first power value of the pixel sequence as a pixel waveform distortion disturbed index of the pixel sequence.
4. The method for nondestructive testing of road and bridge concrete defects according to claim 1, wherein the adaptive wavelet decomposition layer number for further obtaining the pixel sequence comprises the following specific steps:
And marking the rounded value of the pixel waveform distortion disturbed index of the pixel sequence and the minimum value in the first parameter adjusting factor as the self-adaptive wavelet decomposition layer number of the pixel sequence.
5. The method for nondestructive testing of road and bridge concrete defects according to claim 1, wherein the obtaining the concrete integrity of the road and bridge concrete according to the pixel sequence after noise reduction and the pixel sequence in the hyperspectral data of the road and bridge concrete without defect and damage with equal length comprises the following specific steps:
Cosine similarity of pixel sequences in hyperspectral data of road and bridge concrete after noise reduction and pixel sequences in hyperspectral data of road and bridge concrete without defect of equal length is recorded as complete similarity of pixel sequences after noise reduction;
and (3) marking the average value of the complete similarity of all the pixel sequences after noise reduction in all the sampling positions in the hyperspectral data of the road bridge concrete as the concrete integrity of the road bridge concrete.
6. The method for nondestructive testing of road and bridge concrete defects according to claim 1, wherein the method for judging the quality of the road and bridge concrete according to the integrity of the concrete and feeding back to the technician to complete the nondestructive testing of the road and bridge concrete defects comprises the following specific steps:
When the concrete integrity of the road and bridge concrete is greater than or equal to an integrity threshold value, the quality of the road and bridge concrete is considered to be qualified, otherwise, the quality of the road and bridge concrete is considered to be unqualified;
And feeding back the quality judgment result of the road and bridge concrete to engineering acceptance technicians, and realizing nondestructive detection of the road and bridge concrete defects.
7. A road bridge concrete defect non-destructive inspection system comprising a memory, a processor and a computer program stored in said memory and running on said processor, characterized in that said processor, when executing said computer program, implements the steps of the method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410384699.XA CN117969535B (en) | 2024-04-01 | 2024-04-01 | Road bridge concrete defect nondestructive testing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410384699.XA CN117969535B (en) | 2024-04-01 | 2024-04-01 | Road bridge concrete defect nondestructive testing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117969535A true CN117969535A (en) | 2024-05-03 |
| CN117969535B CN117969535B (en) | 2024-06-07 |
Family
ID=90864928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410384699.XA Active CN117969535B (en) | 2024-04-01 | 2024-04-01 | Road bridge concrete defect nondestructive testing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117969535B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105929024A (en) * | 2016-04-21 | 2016-09-07 | 安徽省（水利部淮河水利委员会）水利科学研究院 | Intelligent detection and quantitative recognition method for defect of concrete |
| CN107576619A (en) * | 2017-09-04 | 2018-01-12 | 华北理工大学 | Detect EO-1 hyperion test and the analysis method of concrete erosion product |
| KR20180084479A (en) * | 2017-01-17 | 2018-07-25 | 서울시립대학교 산학협력단 | Apparatus and method for discriminating a concrete status using a hyperspectral image |
| CN112611754A (en) * | 2020-10-26 | 2021-04-06 | 中国水利水电第三工程局有限公司 | Method for evaluating appearance quality of fair-faced concrete |
| KR102254773B1 (en) * | 2020-09-14 | 2021-05-24 | 한국건설기술연구원 | Automatic decision and classification system for each defects of building components using image information, and method for the same |
| KR20230052683A (en) * | 2021-10-13 | 2023-04-20 | 고려대학교 산학협력단 | Evaluation Method of Concrete Compressive Strength Using Hyperspectral Camera and Machine Learning |
| CN117435940A (en) * | 2023-12-20 | 2024-01-23 | 龙建路桥股份有限公司 | Spectrum detection method for winter concrete curing process |
| CN117710380A (en) * | 2024-02-06 | 2024-03-15 | 中九建工集团有限公司 | Intelligent detection method and system for air permeability of concrete |
| CN117788462A (en) * | 2024-02-26 | 2024-03-29 | 湖南铁华精斧汽车集团股份有限公司 | Nondestructive testing method for breakage defect of self-unloading semitrailer axle |
-
2024
- 2024-04-01 CN CN202410384699.XA patent/CN117969535B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105929024A (en) * | 2016-04-21 | 2016-09-07 | 安徽省（水利部淮河水利委员会）水利科学研究院 | Intelligent detection and quantitative recognition method for defect of concrete |
| KR20180084479A (en) * | 2017-01-17 | 2018-07-25 | 서울시립대학교 산학협력단 | Apparatus and method for discriminating a concrete status using a hyperspectral image |
| CN107576619A (en) * | 2017-09-04 | 2018-01-12 | 华北理工大学 | Detect EO-1 hyperion test and the analysis method of concrete erosion product |
| KR102254773B1 (en) * | 2020-09-14 | 2021-05-24 | 한국건설기술연구원 | Automatic decision and classification system for each defects of building components using image information, and method for the same |
| CN112611754A (en) * | 2020-10-26 | 2021-04-06 | 中国水利水电第三工程局有限公司 | Method for evaluating appearance quality of fair-faced concrete |
| KR20230052683A (en) * | 2021-10-13 | 2023-04-20 | 고려대학교 산학협력단 | Evaluation Method of Concrete Compressive Strength Using Hyperspectral Camera and Machine Learning |
| CN117435940A (en) * | 2023-12-20 | 2024-01-23 | 龙建路桥股份有限公司 | Spectrum detection method for winter concrete curing process |
| CN117710380A (en) * | 2024-02-06 | 2024-03-15 | 中九建工集团有限公司 | Intelligent detection method and system for air permeability of concrete |
| CN117788462A (en) * | 2024-02-26 | 2024-03-29 | 湖南铁华精斧汽车集团股份有限公司 | Nondestructive testing method for breakage defect of self-unloading semitrailer axle |
Non-Patent Citations (1)
| Title |
|---|
| 汪金花 等: "混凝土硫酸钠腐蚀产物的高光谱检测方法研究", 《光谱学与光谱分析》, vol. 39, no. 6, 30 June 2019 (2019-06-30), pages 1724 - 1730 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117969535B (en) | 2024-06-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114494210A (en) | Plastic film production defect detection method and system based on image processing | |
| CN113936132B (en) | Method and system for detecting water pollution of chemical plant based on computer vision | |
| CN114782329A (en) | Bearing defect damage degree evaluation method and system based on image processing | |
| CN111879709B (en) | Lake water body spectral reflectivity inspection method and device | |
| CN113129281B (en) | Wheat stem section parameter detection method based on deep learning | |
| CN116958036A (en) | Method for inspecting pollution state of insulator by unmanned aerial vehicle based on hyperspectral technology | |
| CN112098997B (en) | Three-dimensional holographic imaging security inspection radar image foreign matter detection method | |
| CN118225803B (en) | Visual detection method for appearance of blade surface of bulldozer | |
| CN116485779A (en) | Adaptive wafer defect detection method and device, electronic equipment and storage medium | |
| CN115753809A (en) | Insulator contamination detection method, device, equipment and storage medium | |
| CN116721391A (en) | Method for detecting separation effect of raw oil based on computer vision | |
| CN1106572C (en) | Automatic evaluation method of casting ingot cross section quality | |
| CN118334007A (en) | Crack detection and early warning method and system for hydraulic concrete structure | |
| CN117969535B (en) | Road bridge concrete defect nondestructive testing method and system | |
| CN117197048A (en) | Ship water gauge reading detection method, device and equipment | |
| CN113607068B (en) | Method for establishing and extracting recognition model of photoacoustic measurement signal characteristics | |
| CN115761606A (en) | Box electric energy meter identification method and device based on image processing | |
| CN112508862B (en) | Method for enhancing magneto-optical image of crack by improving GAN | |
| CN111507258B (en) | Road area detection method and system based on ensemble learning for focusing key samples | |
| CN114494240A (en) | Ballastless track slab crack measurement method based on multi-scale cooperation deep learning | |
| CN114708190A (en) | Road crack detection and evaluation algorithm based on deep learning | |
| CN112834454A (en) | Winter jujube detection method based on near-infrared hyperspectral technology | |
| CN114252404B (en) | Product authenticity inspection system and method under hyperspectral satellite | |
| CN117474915B (en) | Abnormality detection method, electronic equipment and storage medium | |
| CN114882045B (en) | Technological method for milling casting gate |
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 |