CN111882507A - Metal element identification method and device - Google Patents
Metal element identification method and device Download PDFInfo
- Publication number
- CN111882507A CN111882507A CN202010915111.0A CN202010915111A CN111882507A CN 111882507 A CN111882507 A CN 111882507A CN 202010915111 A CN202010915111 A CN 202010915111A CN 111882507 A CN111882507 A CN 111882507A
- Authority
- CN
- China
- Prior art keywords
- image
- threshold
- selecting
- terahertz
- gray level
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 229910052751 metal Inorganic materials 0.000 title claims abstract description 56
- 239000002184 metal Substances 0.000 title claims abstract description 55
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000000605 extraction Methods 0.000 claims abstract description 30
- 238000007781 pre-processing Methods 0.000 claims abstract description 16
- 230000011218 segmentation Effects 0.000 claims abstract description 15
- 238000003708 edge detection Methods 0.000 claims abstract description 14
- 238000001914 filtration Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 18
- 230000009977 dual effect Effects 0.000 claims description 12
- 230000001629 suppression Effects 0.000 claims description 6
- 238000007689 inspection Methods 0.000 abstract description 17
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 7
- 238000003384 imaging method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000006378 damage Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 229910052755 nonmetal Inorganic materials 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
- G01V8/00—Prospecting or detecting by optical means
- G01V8/10—Detecting, e.g. by using light barriers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Image Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Geophysics (AREA)
- Multimedia (AREA)
- Evolutionary Biology (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Algebra (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a metal element identification method and device. The method comprises the following steps: image preprocessing: collecting a terahertz image of an article, and performing image enhancement, image graying and image filtering operation on the input terahertz image; selecting a threshold value: performing threshold segmentation on the preprocessed terahertz image, and selecting an optimal threshold; feature extraction: applying the obtained optimal threshold value to a canny edge detection algorithm, and performing feature extraction on the preprocessed terahertz image to obtain image feature parameters; and (3) information comparison: and comparing the extracted image characteristic parameters with data in an existing database to judge which metal element the object in the image is. The method has high automation degree, and compared with other image identification methods, the method can more quickly and accurately identify the metal element, particularly the metal cutter, thereby greatly reducing the workload and the false detection rate of manual identification of security personnel and improving the security inspection efficiency.
Description
Technical Field
The invention belongs to the technical field of computer vision, and particularly relates to a metal element identification method and device.
Background
In recent years, in order to ensure public safety, railway stations, airports, subways and event venues are all provided with security inspection systems, and although basic security inspection requirements can be met, the automation degree is low, a large amount of manpower is consumed, and the security inspection efficiency is low. In addition, most of security inspection equipment in the market is an X-ray scanner, but the electronic energy of the security inspection equipment is too high, so that the security inspection equipment can cause ionization injury to human bodies, and therefore the security inspection equipment is only used for detecting objects such as luggage bags and the like. And the human body can only be manually checked, the automation degree is low, and the detection is easy to miss, so that some lawbreakers can carry the cutter to enter public places to fierce and cannot be prevented.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides the metal element identification method and the metal element identification device, which have high automation degree, and compared with other image identification methods, the metal element identification method and the metal element identification device can quickly and accurately identify the metal element, especially the metal cutter, so that the workload and the false inspection rate of manual identification of security personnel are greatly reduced, and the security inspection efficiency is improved.
To achieve the above object, according to one aspect of the present invention, there is provided a metal component identification method including:
image preprocessing: collecting a terahertz image of an article, and performing image enhancement, image graying and image filtering operation on the input terahertz image;
selecting a threshold value: performing threshold segmentation on the preprocessed terahertz image, and selecting an optimal threshold;
feature extraction: applying the obtained optimal threshold value to a canny edge detection algorithm, and performing feature extraction on the preprocessed terahertz image to obtain image feature parameters;
and (3) information comparison: and comparing the extracted image characteristic parameters with data in an existing database to judge which metal element the object in the image is.
In some embodiments, the optimal threshold is chosen using the maximum inter-class variance method.
In some embodiments, selecting the optimal threshold value by using the maximum inter-class variance method specifically includes: counting the number of pixels in each gray level after threshold segmentation, dividing a pixel set, calculating the proportion of the pixel set in each gray level in the whole gray level, obtaining the probability distribution of each pixel in the whole image according to the proportion, traversing the whole gray level, calculating the inter-class probability of the foreground and the background under the current gray level, calculating the threshold value corresponding to the intra-class variance and the inter-class variance, and selecting the gray level with the maximum inter-class variance as the optimal threshold value.
In some embodiments, the image feature parameter is an image edge pixel; the acquiring of the image characteristic parameters comprises:
canny edge detection is carried out on the preprocessed terahertz image, and the gradient amplitude of the image is calculated
And gradient direction θ arctan2 (G)y,Gx) Wherein G isxAs partial derivatives of the image in the x-direction, GyIs the partial derivative of the image in the y direction;
carrying out non-maximum suppression to thin the image edge and keep the local maximum gradient of the image;
and selecting a double threshold according to the optimal threshold, and acquiring image edge pixels by using the double threshold.
In some embodiments, the dual thresholds include a high threshold and a low threshold; selecting the dual threshold according to the optimal threshold comprises: and taking the optimal threshold as a high threshold, and selecting a low threshold within the range of 1/3-1/2 of the high threshold.
In some embodiments, acquiring image edge pixels using dual thresholds comprises: points higher than the high threshold are used as edge point pixels, and points lower than the low threshold are discarded; points greater than the low threshold and less than the high threshold are determined using 8-pass regions, i.e., points greater than the low threshold and less than the high threshold can only be edge points if they are connected to high threshold pixels.
According to another aspect of the present invention, there is provided a metal component recognition apparatus including: the system comprises an image acquisition device, an image processing device and an information comparison device; the input end of the image processing device is connected with the image acquisition device, and the output end of the image processing device is connected with the information comparison device; the image acquisition device is used for acquiring a terahertz image of an article and outputting the terahertz image to the image processing device; the image processing device is used for acquiring characteristic parameters of the terahertz image input by the image acquisition device; the information comparison device is used for comparing the extracted image characteristic parameters with data in an existing database and judging which metal element the object in the image is.
In some embodiments, an image processing apparatus includes an image preprocessing module, a threshold selection module, and a feature extraction module; the input end of the image preprocessing module is connected with the image acquisition device, the output end of the image preprocessing module is respectively connected with the input end of the threshold value selecting module and the first input end of the feature extraction module, the output end of the threshold value selecting module is connected with the second input end of the feature extraction module, and the output end of the feature extraction module is connected with the information comparison device;
the image preprocessing module is used for performing image enhancement, image graying and image filtering operations on the input terahertz image and outputting the preprocessed image to the threshold selecting module and the feature extracting module; the threshold selection module is used for carrying out threshold segmentation on the preprocessed terahertz image, selecting an optimal threshold and outputting the optimal threshold to the feature extraction module; the feature extraction module is used for applying the obtained optimal threshold value to a canny edge detection algorithm, performing feature extraction on the preprocessed terahertz image, and obtaining image feature parameters.
In some embodiments, the selecting the optimal threshold value by using the maximum inter-class variance method specifically includes: counting the number of pixels in each gray level after threshold segmentation, dividing a pixel set, calculating the proportion of the pixel set in each gray level in the whole gray level, obtaining the probability distribution of each pixel in the whole image according to the proportion, traversing the whole gray level, calculating the inter-class probability of the foreground and the background under the current gray level, calculating the threshold value corresponding to the intra-class variance and the inter-class variance, and selecting the gray level with the maximum inter-class variance as the optimal threshold value.
In some embodiments, the image feature parameter is an image edge pixel; the acquiring of the image characteristic parameters comprises:
canny edge detection is carried out on the preprocessed terahertz image, and the gradient amplitude of the image is calculated
And gradient direction θ arctan2 (G)y,Gx) Wherein G isxAs partial derivatives of the image in the x-direction, GyIs the partial derivative of the image in the y direction;
carrying out non-maximum suppression to thin the image edge and keep the local maximum gradient of the image;
and selecting a double threshold according to the optimal threshold, and acquiring image edge pixels by using the double threshold.
Generally, compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects: the terahertz imaging technology is combined with the image recognition technology to obtain the information of the target object, the information is compared with the data of the database, and the type of the metal element of the target object, especially the type of the metal cutter, can be accurately recognized. The invention can be widely applied to the field of security inspection, greatly reduces the workload of manual identification and the false inspection rate of security inspection personnel, and improves the security inspection efficiency.
Drawings
FIG. 1 is a view showing the construction of a metal component recognition apparatus according to an embodiment of the present invention;
FIG. 2 is a flow chart of a metal component identification method according to an embodiment of the present invention;
fig. 3 is a flow chart of the canny edge detection algorithm according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
As shown in fig. 1, a metal element identification device according to an embodiment of the present invention includes: the device comprises an image acquisition device, an image processing device and an information comparison device. The input end of the image processing device is connected with the image acquisition device, and the output end of the image processing device is connected with the information comparison device. The image acquisition device is used for acquiring terahertz images of articles and outputting the terahertz images to the image processing device. The image processing device is used for acquiring characteristic parameters of the terahertz image input by the image acquisition device. The information comparison device is used for comparing the extracted image characteristic parameters with data in an existing database and judging which metal element the object in the image is.
In some embodiments, the information comparing device obtains an overall profile of the article by connecting edge pixels of the image, and compares the overall profile with an article profile in an existing database to determine what kind of metal element the article in the image is. In some embodiments, the database is a metal tool feature database. In some embodiments, the information comparing device compares the extracted image characteristic parameters with data in an existing metal tool characteristic database to determine whether the object in the image is a metal tool. In some embodiments, the information comparing device determines whether the object in the image is a metal tool according to the similarity between the extracted image feature parameter and the data in the existing metal tool feature database. In some embodiments, the information comparison device determines that the object in the image is a metal tool when the similarity is greater than 85%. In some embodiments, when the information comparing device determines that the article in the image is a metal cutter, it further determines what type of metal cutter the article in the image is.
In some embodiments, an image processing apparatus includes an image preprocessing module, a threshold extraction module, and a feature extraction module. The input end of the image preprocessing module is connected with the image acquisition device, the output end of the image preprocessing module is respectively connected with the input end of the threshold selecting module and the first input end of the feature extraction module, the output end of the threshold selecting module is connected with the second input end of the feature extraction module, and the output end of the feature extraction module is connected with the information comparison device.
In some embodiments, the image preprocessing module is configured to perform image enhancement, image graying and image filtering operations on the input terahertz image, and output the preprocessed image to the threshold selection module and the feature extraction module. In some embodiments, image enhancement comprises: and the contrast ratio of the terahertz image target and the background is improved by using a Gamma correction method. In some embodiments, the image graying comprises: calculating gray value of terahertz image
And obtaining the required gray-scale image, wherein R is a red value, G is a green value, and B is a blue value. In some embodiments, the image filtering comprises: and a high-pass filtering method is used for enhancing the high-frequency signal at the edge of the terahertz image, so that the outline of the terahertz image is clearer.
In some embodiments, the threshold selection module is configured to perform threshold segmentation on the preprocessed terahertz image, select an optimal threshold, and output the optimal threshold to the feature extraction module. In some embodiments, the optimal threshold is chosen using the variance between the largest classes. In some embodiments, selecting the optimal threshold value by using the variance method between the maximum classes specifically includes: counting the number of pixels in each gray level after threshold segmentation, dividing a pixel set, calculating the proportion of the pixel set in each gray level in the whole gray level, obtaining the probability distribution of each pixel in the whole image according to the proportion, traversing the whole gray level, calculating the inter-class probability of the foreground and the background under the current gray level, calculating the threshold value corresponding to the intra-class variance and the inter-class variance, and selecting the gray level with the maximum inter-class variance as the optimal threshold value. In some embodiments, a histogram is used for analysis, the number of pixels in each gray level after threshold segmentation is counted, the preprocessed terahertz image is divided into a plurality of (for example, 240) small blocks, a hidden region of the tool is found, the proportion of a pixel set in each gray level in the whole gray level (for example, [ 0255 ]) is calculated, and an optimal threshold is selected.
In some embodiments, the feature extraction module is configured to apply the obtained optimal threshold to a canny edge detection algorithm, perform feature extraction on the preprocessed terahertz image, and obtain an image feature parameter. In some embodiments, the image feature parameter is an image edge pixel. In some embodiments, obtaining image feature parameters comprises: canny edge detection is carried out on the preprocessed terahertz image, and the gradient amplitude of the image is calculated
And gradient direction θ arctan2 (G)y,Gx) Wherein G isxAs partial derivatives of the image in the x-direction, GyIs the partial derivative of the image in the y direction; carrying out non-maximum suppression to thin the image edge and keep the local maximum gradient of the image; and selecting a double threshold according to the optimal threshold, and acquiring image edge pixels by using the double threshold. In some embodiments, the dual threshold includes a high threshold and a low threshold. In some embodiments, selecting the dual threshold according to the optimal threshold comprises: and taking the optimal threshold as a high threshold, and selecting a low threshold within the range of 1/3-1/2 of the high threshold. In some embodiments, acquiring image edge pixels using dual thresholds comprises: points higher than the high threshold are used as edge point pixels, and points lower than the low threshold are discarded; points greater than the low threshold and less than the high threshold are determined using 8-pass regions, i.e., points greater than the low threshold and less than the high threshold can only be edge points if they are connected to high threshold pixels.
As shown in fig. 2, the metal element identification method according to the embodiment of the present invention mainly includes the following steps:
(1) image preprocessing: collecting a terahertz image of an article from a terahertz system, and performing image enhancement, image graying and image filtering operation on the input terahertz image;
in some embodiments, image enhancement comprises: and the contrast ratio of the terahertz image target and the background is improved by using a Gamma correction method.
In some embodiments, the image graying comprises: and (3) calculating the Gray value Gray of the terahertz image by using a formula (1) to obtain the required Gray image.
Wherein R is a red color value, G is a green color value, and B is a blue color value.
In some embodiments, the image filtering comprises: and a high-pass filtering method is used for enhancing the high-frequency signal at the edge of the terahertz image, so that the outline of the terahertz image is clearer.
(2) Selecting a threshold value: performing threshold segmentation on the preprocessed terahertz image, and selecting an optimal threshold;
in some embodiments, the optimal threshold is chosen using the variance between the largest classes.
In some embodiments, selecting the optimal threshold value by using the variance method between the maximum classes specifically includes: counting the number of pixels in each gray level after threshold segmentation, dividing a pixel set, calculating the proportion of the pixel set in each gray level in the whole gray level, obtaining the probability distribution of each pixel in the whole image according to the proportion, traversing the whole gray level, calculating the inter-class probability of the foreground and the background under the current gray level, calculating the threshold value corresponding to the intra-class variance and the inter-class variance, and selecting the gray level with the maximum inter-class variance as the optimal threshold value.
In some embodiments, a histogram is used for analysis, the number of pixels in each gray level after threshold segmentation is counted, the preprocessed terahertz image is divided into a plurality of (for example, 240) small blocks, a hidden region of the tool is found, the proportion of a pixel set in each gray level in the whole gray level (for example, [ 0255 ]) is calculated, and an optimal threshold is selected.
(3) Feature extraction: applying the obtained optimal threshold value to a canny edge detection algorithm, and performing feature extraction on the preprocessed terahertz image to obtain image feature parameters, as shown in fig. 3;
in some embodiments, the image feature parameter is an image edge pixel.
In some embodiments, obtaining image feature parameters comprises:
carrying out canny edge detection on the preprocessed terahertz image, and calculating the gradient amplitude G and the gradient direction theta of the image by using formulas (2) and (3);
θ＝arctan2(Gy,Gx) (3)
wherein G isxAs partial derivatives of the image in the x-direction, GyIs the partial derivative of the image in the y-direction.
Carrying out non-maximum suppression to thin the image edge and keep the local maximum gradient of the image;
and selecting a double threshold according to the optimal threshold, and acquiring image edge pixels by using the double threshold.
In some embodiments, the dual threshold includes a high threshold and a low threshold.
In some embodiments, selecting the dual threshold according to the optimal threshold comprises: and taking the optimal threshold as a high threshold, and selecting a low threshold within the range of 1/3-1/2 of the high threshold.
In some embodiments, acquiring image edge pixels using dual thresholds comprises: points higher than the high threshold are used as edge point pixels, and points lower than the low threshold are discarded; points greater than the low threshold and less than the high threshold are determined using 8-pass regions, i.e., points greater than the low threshold and less than the high threshold can only be edge points if they are connected to high threshold pixels.
(4) And (3) information comparison: and comparing the extracted image characteristic parameters with data in an existing database to judge which metal element the object in the image is.
In some embodiments, the information alignment comprises: and connecting the image edge pixels to obtain the overall outline of the article, comparing the overall outline with the outline of the article in the existing database, and judging which metal element the article in the image is.
In some embodiments, the database is a metal tool feature database.
In some embodiments, the extracted image characteristic parameters are compared with data in an existing metal tool characteristic database to determine whether the object in the image is a metal tool.
In some embodiments, whether the object in the image is a metal tool is determined according to the similarity between the extracted image feature parameters and the data in the existing metal tool feature database.
In some embodiments, when the similarity is greater than 85%, the article in the image is judged to be a metal cutter.
In some embodiments, determining that the article in the image is a metal tool further comprises determining what type of metal tool the article in the image is.
At present, the development of terahertz imaging technology is more and more mature, and the terahertz imaging technology has the advantage of high imaging resolution. The frequency of terahertz radiation is between 0.1 and 100Thz, the terahertz radiation cannot penetrate through metal objects, but can penetrate through various non-metal objects, and no harm is caused to human bodies, so that the terahertz imaging technology has great advantage in being applied to a security inspection system. The terahertz security inspection system utilizes the characteristics of the terahertz imaging technology and is matched with the mature image recognition technology, so that the terahertz security inspection is more and more automated.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A metal component identification method, comprising:
image preprocessing: collecting a terahertz image of an article, and performing image enhancement, image graying and image filtering operation on the input terahertz image;
selecting a threshold value: performing threshold segmentation on the preprocessed terahertz image, and selecting an optimal threshold;
feature extraction: applying the obtained optimal threshold value to a canny edge detection algorithm, and performing feature extraction on the preprocessed terahertz image to obtain image feature parameters;
and (3) information comparison: and comparing the extracted image characteristic parameters with data in an existing database to judge which metal element the object in the image is.
2. A method for identifying a metal component as claimed in claim 1, wherein the optimum threshold is selected using the method of variance between maximum classes.
3. A method for identifying a metallic element as in claim 2, wherein selecting the optimal threshold value using the maximum inter-class variance method comprises: counting the number of pixels in each gray level after threshold segmentation, dividing a pixel set, calculating the proportion of the pixel set in each gray level in the whole gray level, obtaining the probability distribution of each pixel in the whole image according to the proportion, traversing the whole gray level, calculating the inter-class probability of the foreground and the background under the current gray level, calculating the threshold value corresponding to the intra-class variance and the inter-class variance, and selecting the gray level with the maximum inter-class variance as the optimal threshold value.
4. The metal element identification method of claim 1, wherein the image characteristic parameter is an image edge pixel; the acquiring of the image characteristic parameters comprises:
canny edge detection is carried out on the preprocessed terahertz image, and the gradient amplitude of the image is calculated
And gradient direction θ arctan2 (G)y,Gx) Wherein G isxAs partial derivatives of the image in the x-direction, GyAs partial derivatives of the image in the y-direction；
Carrying out non-maximum suppression to thin the image edge and keep the local maximum gradient of the image;
and selecting a double threshold according to the optimal threshold, and acquiring image edge pixels by using the double threshold.
5. The metal element identification method of claim 4, wherein the dual threshold comprises a high threshold and a low threshold; selecting the dual threshold according to the optimal threshold comprises: and taking the optimal threshold as a high threshold, and selecting a low threshold within the range of 1/3-1/2 of the high threshold.
6. The metal element identification method of claim 4, wherein acquiring image edge pixels using a dual threshold comprises: points higher than the high threshold are used as edge point pixels, and points lower than the low threshold are discarded; points greater than the low threshold and less than the high threshold are determined using 8-pass regions, i.e., points greater than the low threshold and less than the high threshold can only be edge points if they are connected to high threshold pixels.
7. A metal component recognition apparatus, comprising: the system comprises an image acquisition device, an image processing device and an information comparison device; the input end of the image processing device is connected with the image acquisition device, and the output end of the image processing device is connected with the information comparison device; the image acquisition device is used for acquiring a terahertz image of an article and outputting the terahertz image to the image processing device; the image processing device is used for acquiring characteristic parameters of the terahertz image input by the image acquisition device; the information comparison device is used for comparing the extracted image characteristic parameters with data in an existing database and judging which metal element the object in the image is.
8. The metal component identifying device according to claim 7, wherein the image processing device includes an image preprocessing module, a threshold selecting module, and a feature extracting module; the input end of the image preprocessing module is connected with the image acquisition device, the output end of the image preprocessing module is respectively connected with the input end of the threshold value selecting module and the first input end of the feature extraction module, the output end of the threshold value selecting module is connected with the second input end of the feature extraction module, and the output end of the feature extraction module is connected with the information comparison device;
the image preprocessing module is used for performing image enhancement, image graying and image filtering operations on the input terahertz image and outputting the preprocessed image to the threshold selecting module and the feature extracting module; the threshold selection module is used for carrying out threshold segmentation on the preprocessed terahertz image, selecting an optimal threshold and outputting the optimal threshold to the feature extraction module; the feature extraction module is used for applying the obtained optimal threshold value to a canny edge detection algorithm, performing feature extraction on the preprocessed terahertz image, and obtaining image feature parameters.
9. The metal element identification device of claim 8, wherein the selecting the optimal threshold value using the maximum inter-class variance method comprises: counting the number of pixels in each gray level after threshold segmentation, dividing a pixel set, calculating the proportion of the pixel set in each gray level in the whole gray level, obtaining the probability distribution of each pixel in the whole image according to the proportion, traversing the whole gray level, calculating the inter-class probability of the foreground and the background under the current gray level, calculating the threshold value corresponding to the intra-class variance and the inter-class variance, and selecting the gray level with the maximum inter-class variance as the optimal threshold value.
10. The metal element recognition apparatus of claim 8, wherein the image characteristic parameter is an image edge pixel; the acquiring of the image characteristic parameters comprises:
canny edge detection is carried out on the preprocessed terahertz image, and the gradient amplitude of the image is calculated
And gradient direction θ arctan2 (G)y,Gx) Wherein G isxAs partial derivatives of the image in the x-direction, GyIs the partial derivative of the image in the y direction;
carrying out non-maximum suppression to thin the image edge and keep the local maximum gradient of the image;
and selecting a double threshold according to the optimal threshold, and acquiring image edge pixels by using the double threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010915111.0A CN111882507A (en) | 2020-09-03 | 2020-09-03 | Metal element identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010915111.0A CN111882507A (en) | 2020-09-03 | 2020-09-03 | Metal element identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111882507A true CN111882507A (en) | 2020-11-03 |
Family
ID=73199564
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010915111.0A Pending CN111882507A (en) | 2020-09-03 | 2020-09-03 | Metal element identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111882507A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117409225A (en) * | 2023-10-31 | 2024-01-16 | 深圳市大满包装有限公司 | Intelligent metal tank sensing method and system based on man-machine interaction |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107356599A (en) * | 2017-06-23 | 2017-11-17 | 厦门大学 | A kind of Terahertz lossless detection method of ceramic matric composite |
| CN108764328A (en) * | 2018-05-24 | 2018-11-06 | 广东工业大学 | The recognition methods of Terahertz image dangerous material, device, equipment and readable storage medium storing program for executing |
| CN108898132A (en) * | 2018-05-25 | 2018-11-27 | 广东工业大学 | A kind of terahertz image dangerous material recognition methods based on Shape context description |
| CN109447036A (en) * | 2018-11-16 | 2019-03-08 | 齐鲁工业大学 | A kind of segmentation of image digitization and recognition methods and system |
| CN110009653A (en) * | 2019-03-12 | 2019-07-12 | 江苏理工学院 | Increase limb recognition point sharp picture based on gray level threshold segmentation method and knows method for distinguishing |
| CN212846888U (en) * | 2020-09-03 | 2021-03-30 | 浙江长芯光电科技有限公司 | Metal element recognition device |
-
2020
- 2020-09-03 CN CN202010915111.0A patent/CN111882507A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107356599A (en) * | 2017-06-23 | 2017-11-17 | 厦门大学 | A kind of Terahertz lossless detection method of ceramic matric composite |
| CN108764328A (en) * | 2018-05-24 | 2018-11-06 | 广东工业大学 | The recognition methods of Terahertz image dangerous material, device, equipment and readable storage medium storing program for executing |
| CN108898132A (en) * | 2018-05-25 | 2018-11-27 | 广东工业大学 | A kind of terahertz image dangerous material recognition methods based on Shape context description |
| CN109447036A (en) * | 2018-11-16 | 2019-03-08 | 齐鲁工业大学 | A kind of segmentation of image digitization and recognition methods and system |
| CN110009653A (en) * | 2019-03-12 | 2019-07-12 | 江苏理工学院 | Increase limb recognition point sharp picture based on gray level threshold segmentation method and knows method for distinguishing |
| CN212846888U (en) * | 2020-09-03 | 2021-03-30 | 浙江长芯光电科技有限公司 | Metal element recognition device |
Non-Patent Citations (1)
| Title |
|---|
| 王嘉俊等: "改进Canny算子在水面目标边缘检测中的研究", 《计算机时代》, 31 January 2020 (2020-01-31), pages 35 - 38 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117409225A (en) * | 2023-10-31 | 2024-01-16 | 深圳市大满包装有限公司 | Intelligent metal tank sensing method and system based on man-machine interaction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110334706B (en) | Image target identification method and device | |
| Sedighi et al. | A new and robust method for character segmentation and recognition in license plate images | |
| CN102364496B (en) | Method and system for identifying automobile license plates automatically based on image analysis | |
| CN111260616A (en) | Insulator crack detection method based on Canny operator two-dimensional threshold segmentation optimization | |
| CN103077407B (en) | Car logo positioning and recognition method and car logo positioning and recognition system | |
| CN109409355B (en) | Novel transformer nameplate identification method and device | |
| CN107389701A (en) | A kind of PCB visual defects automatic checkout system and method based on image | |
| CN103824091B (en) | A kind of licence plate recognition method for intelligent transportation system | |
| CN110210477B (en) | Digital instrument reading identification method | |
| CN105678213B (en) | Dual-mode mask person event automatic detection method based on video feature statistics | |
| US9443160B2 (en) | Irradiation field recognition | |
| CN108898132A (en) | A kind of terahertz image dangerous material recognition methods based on Shape context description | |
| Prabhakar et al. | A novel design for vehicle license plate detection and recognition | |
| CN104408780A (en) | Face recognition attendance system | |
| CN109102003A (en) | A kind of small target detecting method and system based on Infrared Physics Fusion Features | |
| CN106650623A (en) | Face detection-based method for verifying personnel and identity document for exit and entry | |
| CN111027544B (en) | MSER license plate positioning method and system based on visual saliency detection | |
| CN114972325A (en) | Automobile hub defect detection method based on image processing | |
| CN118279304B (en) | Abnormal recognition method, device and medium for special-shaped metal piece based on image processing | |
| CN118038515B (en) | Face recognition method | |
| Paunwala et al. | A novel multiple license plate extraction technique for complex background in Indian traffic conditions | |
| CN117036259A (en) | Metal plate surface defect detection method based on deep learning | |
| CN112288682A (en) | Electric power equipment defect positioning method based on image registration | |
| CN110321890B (en) | Digital instrument identification method of power inspection robot | |
| CN212846888U (en) | Metal element recognition device |
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 |