CN113269745A - Aerial photography automobile counting method based on OpenCv - Google Patents
Aerial photography automobile counting method based on OpenCv Download PDFInfo
- Publication number
- CN113269745A CN113269745A CN202110560068.5A CN202110560068A CN113269745A CN 113269745 A CN113269745 A CN 113269745A CN 202110560068 A CN202110560068 A CN 202110560068A CN 113269745 A CN113269745 A CN 113269745A
- Authority
- CN
- China
- Prior art keywords
- image
- automobile
- processing
- aerial vehicle
- opencv
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 38
- 238000001931 thermography Methods 0.000 claims abstract description 16
- 238000000605 extraction Methods 0.000 claims abstract description 4
- 230000000877 morphologic effect Effects 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims description 12
- 238000009499 grossing Methods 0.000 claims description 11
- 239000003086 colorant Substances 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 abstract description 7
- 239000000463 material Substances 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 27
- 238000006243 chemical reaction Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000010339 dilation Effects 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 1
- 230000036760 body temperature Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000005530 etching Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000003137 locomotive effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Abstract
The invention provides an aerial photography automobile counting method based on OpenCv, and relates to the technical field of computer vision. The method comprises the steps of firstly, aiming at the shadow in the image, eliminating the influence of the shadow on counting by a method of adjusting contrast; then, highlighting automobile characteristics by using a thermal imaging model and carrying out gray level processing; then adjusting parameters of the adaptive threshold value, and carrying out binarization processing on the image; then, performing expansion, closing and re-expansion processing on the image by utilizing expansion and closing operations in morphological operations; and finally, calculating the number of connected domains in the image by contour extraction and connected domain counting so as to obtain the number of the automobiles. The method and the device can efficiently and quickly calculate the number of the automobiles, thereby saving manpower and material resources. Under the condition of higher automobile density, the obtained result is still more accurate, and the counting accuracy is improved.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to an aerial photography automobile counting method based on OpenCv.
Background
The automobile is an important transportation tool for people to travel instead of walk and travel in daily life in the current society. With the rapid development of economy in China and the improvement of living standard of people, one automobile often cannot meet the requirement of one family, and even two or more automobiles can appear. With the increase of the demand, the import of the automobile is a solution for relieving the demand, and after the automobile is imported, the automobile needs to be parked in a parking lot for checking. Therefore, it is a strict procedure to specify the number of cars. Common ways to meter the number of locomotives: one is to measure manually, for the cars with a small number of imports, the accuracy of manual measurement is high, but when the cars are in a large number, the problems of low efficiency, high error rate, high labor cost and the like can occur through manual measurement; one is to install a metering device and calculate by driving in a vehicle, but this method is not convenient enough for counting cars, and the metering device needs to be moved at any time, thereby causing waste of time.
Computer vision is a technology that has emerged in the mid-20 th century. With the development of computer vision, the problems of slow code research, instability, independence and incompatibility with other libraries, high business cost, special solutions depending on some hardware and the like occur in the computer vision. In the face of the problems, OpenCv provides a rich visual processing algorithm, optimized C language writing improves the execution speed, and in addition, the open source characteristic of OpenCv enables a link generation execution program to be completely compiled without adding new external support. And the code of OpenCv is easy to run in DSP system and ARM embedded system. OpenCv can run on a plurality of platforms such as Windows, Andriod, Ios, Linux and the like, and is widely applied to the fields of human-computer interaction, virtual reality, face recognition, machine vision and the like. Especially in the aspects of measurement and control measurement, such as automobile distance measurement, body temperature monitoring, automobile counting and the like. In the counting and shooting process, the shielding objects, the refraction and the reflection of light, misoperation and the like in a shooting field all influence the counting precision, factors such as image noise also influence the automobile counting, and then larger errors occur.
Disclosure of Invention
The technical problem to be solved by the invention is to provide an aerial photography automobile counting method based on OpenCv, the method can efficiently and quickly calculate the number of automobiles, so that manpower and material resources are saved, the obtained result is still more accurate under the condition of high automobile density, and the counting accuracy is improved. .
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
an aerial photography automobile counting method based on OpenCv is characterized in that: the method comprises the following steps:
step 1), shooting a parking lot area by using an unmanned aerial vehicle to obtain an automobile image;
step 2), adjusting the contrast of the automobile image obtained in the step 1), performing Gaussian smoothing on the image, processing the image by using a thermal imaging model, performing gray value transformation on the obtained image, and performing self-adaptive binarization processing on the gray image;
step 3), performing morphological operation on the image obtained in the step 2), processing the image by using a binary function through an expansion function, and processing the image by using a closed operation;
step 4), operating the picture obtained in the step 3) by using an inverse function, and processing the image by using an expansion function;
and 5) carrying out contour extraction on the image obtained in the step 4), and calculating connected domains so as to obtain the number of the automobiles.
The technical scheme of the invention is further improved as follows: in the step 1), the height of the unmanned aerial vehicle is kept between 40 meters and 80 meters in the process of aerial photography of the unmanned aerial vehicle; meanwhile, the included angle between the shooting angle of the camera of the unmanned aerial vehicle and the vertical direction is kept within the range of 20 degrees.
The technical scheme of the invention is further improved as follows: and 3) processing the image by using convolution operation in the expansion function in the step 3), wherein the image is adjusted by the self-adaptive binarization function, three channels B, G, R of the background image are 255, and three channel values of the target image are 0.
The technical scheme of the invention is further improved as follows: after the black area is eliminated in the step 4), the small black spots still exist in the image, and in order to further eliminate the small black spots, the whole image is subjected to negation operation, at this time, black is taken as a background, white is taken as an automobile target, and the white automobile target is further expanded by further utilizing an expansion function.
The technical scheme of the invention is further improved as follows: and 5) setting a container, traversing all image points of the whole image, finding a white connected region, marking the connected region, adding random colors for distinguishing, and finally displaying the number of the color regions in the image.
Due to the adoption of the technical scheme, the invention has the technical progress that:
according to the invention, by combining unmanned aerial vehicle aerial photography and OpenCV, the quantity of the automobiles can be efficiently and rapidly calculated, so that manpower and material resources are saved. Meanwhile, under the condition of higher automobile density, the obtained result is still more accurate, and the counting accuracy is improved. Aiming at the condition of high automobile density, the invention realizes the distance conversion between automobile targets by utilizing secondary expansion; meanwhile, the thermal imaging model is utilized, so that the characteristics of the automobile are obviously changed, and the subsequent counting precision is improved; for the problems of noise, black spots and the like in the image, the method utilizes Gaussian smoothing, primary expansion and closing operation to obtain good inhibition and processing.
Unmanned aerial vehicle carries out the in-process of taking photo by plane, and the height that keeps unmanned aerial vehicle can enough guarantee the definition of image for between 40 meters to 80 meters, can guarantee again that the parking area car is as far as possible in camera field of vision to guarantee the accuracy of count.
In the step 2), in the shooting process of the unmanned aerial vehicle, the shooting is influenced by factors such as the intensity of outside illumination, the shielding of ground obstacles, the reflection of the surface of an object to light and the like, and the influence of factors such as refraction and reflection of light is weakened by adjusting the contrast of the image; meanwhile, the thermal imaging model is used for processing the image, so that the characteristics of the automobile can be displayed, and the counting of the automobile is further improved; in the aerial image of the automobile, the automobile is taken as an object to be processed, the proportion of the automobile in the image is large, the image noise can be reduced by utilizing a Gaussian smooth function, and the small details in the image are reduced; and (3) carrying out self-adaptive binarization on the image after Gaussian smoothing, wherein compared with a common binarization function, the self-adaptive binarization function can distinguish the background from the target of the image according to the gray characteristic.
The disconnected regions in the black region can be eliminated through the expansion function in the step 3).
After the black area is eliminated in the step 4), the small black spots still exist in the image, and in order to further eliminate the small black spots, the whole image is subjected to negation operation, at this time, black is taken as a background, white is taken as an automobile target, and the expansion function is further utilized to further expand the white automobile target and eliminate the influence of the small white noise points on the counting.
Drawings
FIG. 1 is an image grayed out after thermal imaging;
FIG. 2 is a grayed adaptive binarized image;
FIG. 3 is a dilated image after binarization;
FIG. 4 is an image of a close operation after dilation processing;
FIG. 5 is an inverted image after a close operation;
FIG. 6 is a second, inverted, dilated image;
FIG. 7 is an image drawn after contour extraction;
FIG. 8 is a calculated number of cars image;
fig. 9 is an overall flowchart.
Detailed Description
The present invention will be described in further detail with reference to the following examples:
as shown in fig. 1 to 9, an OpenCv-based aerial vehicle counting method includes the following steps:
step 1), shooting a parking lot area by using an unmanned aerial vehicle to obtain an automobile image, and keeping the height of the unmanned aerial vehicle between 40 meters and 80 meters in the process of aerial shooting by the unmanned aerial vehicle; meanwhile, the included angle between the shooting angle of the camera of the unmanned aerial vehicle and the vertical direction is kept within the range of 20 degrees;
step 2), adjusting the contrast of the automobile image obtained in the step 1), performing Gaussian smoothing on the image, processing the image by using a thermal imaging model, performing gray value transformation on the obtained image, and performing self-adaptive binarization processing on the gray image; unmanned aerial vehicle can be because factors such as the shading of external light intensity, ground barrier and the reflection of object surface to light influence the shooting at the in-process of shooing, through carrying out contrast adjustment to the image, weakens the influence of factors such as refraction, reflection of light. In the shooting process, the shadow of the automobile exists in most cases, the image is reinforced to a certain degree through adjusting the image contrast, the shadow of the automobile can be lightened, and the counting precision can be improved; in the image, the automobile is used as an object to be processed, the thermal imaging model processes the image, the characteristics of the automobile can be displayed, and the counting of the automobile is further improved; in an automobile aerial image, an automobile is taken as an object to be processed, the proportion of the automobile in the image is the largest, so that the weight occupied by an automobile pixel in an original pixel is the largest, and Gaussian smoothing can equally divide the adjacent pixels of each pixel in a weighted manner, so that image noise can be reduced by using a Gaussian smoothing function, small details in the image can be reduced, the condition that the automobile part appears at the edge of the image always occurs in the image, and the image edge can be better reserved by using the Gaussian smoothing function; the image after Gaussian smoothing is subjected to adaptive binarization, the situation of large enhancement degree can occur in the process of enhancing the contrast of the image, a common binarization function is often subjected to binarization processing only for a class of images, if special situations occur, the binarization processing effect is poor, and the adaptive binarization function can distinguish the background from the target of the image according to the gray characteristic; because the thermal imaging model is used for processing the image, the image is shot within half an hour after the sun is set in the evening, the heat source is clear and easy to identify;
step 3), performing morphological operation on the image obtained in the step 2), processing the image by using an expansion function and a binarization function, processing the image by using a closed operation, and processing the image by using convolution operation in the expansion function, wherein at the moment, the image is adjusted by using a self-adaptive binarization function, three channels B, G, R of the background image are all 255, and the value of three channels of the target image is 0; after the image is subjected to self-adaptive binarization processing, the values of the background pixels three channels B, G, R are all 255, and the values of the target functions three channels are 0. After the contrast ratio, the thermal imaging model processing and the Gaussian smoothing are adjusted, partial residues of shadows, shields and the like still exist in the binary image, and if the residues are not processed, the subsequent counting result is influenced. In the binarized image, the background part is white, and by performing the expansion processing on the background part by using the convolution operation, the white background covers the black area with smaller connected domain in the binarized image, and the larger black connected domain is covered by a part of the same way, but is not completely covered. In order to restore the black connected domain containing the effective information, the image is processed by closing operation again, and the closing operation is specifically implemented by performing expansion processing on the image and then performing corrosion processing. After expansion processing, the black connected domain originally containing effective information can be recovered;
step 4), operating the picture obtained in the step 3) by using an inverse function, processing the image by using an expansion function, removing black areas, and performing inverse operation on the whole image, wherein black is a background and white is an automobile target, and the expansion function is further used for further expanding the white automobile target; after the etching operation and the closing operation, fewer small black spots still exist to influence the counting precision, and at the moment, the binary image is subjected to the negation operation, namely the white background area is changed into the black background, and the black connected area is changed into the white connected area. Performing expansion processing on the inverted image again, wherein the small white connected domain spots are less and even completely eliminated;
and 5) extracting the outline of the image obtained in the step 4), calculating connected domains, further obtaining the number of automobiles, setting containers, traversing all image points of the whole image, further finding a white connected domain, marking the connected domain, adding random colors for distinguishing, and finally displaying the number of colored domains in the image.
The invention carries out aerial photography on the automobile in the parking area by using the unmanned aerial vehicle to carry the camera, and carries out image processing by using the Visual Studio to carry the OpenCv Visual library.
The implementation platform of the invention is an Intel (R) CORE (TM) I5-10500 processor, the main frequency is 3.1GHz, the memory is 8GB, the software platform is a Win 1064 bit operating system, and OpenCV3.4.1 is carried by Visual Studio 2017. The specific implementation mode comprises the following steps:
in the process of aerial photography, the unmanned aerial vehicle should shoot in a time period with abundant sunlight as far as possible, the light intensity is selected within the range of 60000 lx-100000 lx in a sunny day, and the light intensity is selected above 10000lx in a cloudy day, so that the aerial photography effect of the unmanned aerial vehicle is better, and the influence of vehicle shadows on automobile counting can be reduced as far as possible; when the unmanned aerial vehicle is controlled to shoot, the camera is vertical to the ground or the included angle between the camera and the ground vertical line is kept within the range of 15 degrees; in the process of aerial photography, a time period with abundant illumination is selected as much as possible for shooting, after the sun inclines to a certain degree, a large-area shadow appears on an automobile, and if the distance between the automobiles is short, the shadows overlap with each other, so that the shooting effect is influenced; if the light is too strong, objects which are easy to reflect, such as automobile glass, automobile covers, rearview mirrors and the like, can also affect the shooting effect. If the number of the automobiles is small, the selected height can be about 40 meters, and if the automobiles in the parking lot are dense and the number of the automobiles in the parking lot is large, the height of the unmanned aerial vehicle is kept about 80 meters; when the angle and the height of the unmanned aerial vehicle are well controlled, the unmanned aerial vehicle selects to shoot in the central zone of the parking area.
Fig. 1 and 2 show preprocessing of an image, before contrast adjustment of the image, a blank image with the same size as that of an aerial image needs to be created, each pixel of the aerial image is traversed, because the aerial image is a color image, the value of B, G, R three channels in each pixel needs to be obtained, and then a formula is used for obtaining the value of B, G, R three channels
Adjusting each pixel, wherein
And β are gain and offset, respectively; in an aerial image, the image has noise, meanwhile, because other tiny obstacles can influence the result, the image is processed by Gaussian smoothing, firstly, the image pixels after the contrast adjustment are sampled, then, the shape of a Gaussian kernel is determined, the automobile is aerial-photographed, the top view of the automobile can be seen as a rectangle, therefore, the shape of the Gaussian kernel is determined to be the rectangular shape, and then, the Gaussian kernel and the image after the contrast adjustment are used for carrying out convolution operation; although OpenCV can accurately identify a gray level image and further obtain a gray level value of 256, the color change of the image is more accurate, a thermal imaging model is established, thermal imaging change is carried out on the image after Gaussian blur, the image is firstly indexed, and two matrixes are obtained: the method comprises the steps of establishing a thermal imaging model by utilizing a COLORMAP _ HOT color chart to finally obtain a thermal imaging pseudo color chart, wherein the data matrix and the color mapping matrix are used for storing colors in a color image, and the color chart is set in 12 in OpenCV; in order to further highlight the characteristics of an automobile and the characteristics which are not needed after thermal imaging all the time, the gray scale conversion is carried out on the color thermal imaging image, compared with the color image in which the number of pixel values is in an exponential growth type, the number of the pixel values in the image after the gray scale conversion is 256, the time of the later-stage whole image indexing is effectively shortened, and in the gray scale conversion process, B, G, R three channels of each pixel are set with different weights, namely gray is 0.144B +0.587G + 0.299R; after the image is subjected to graying conversion, the image is not black and white which can be seen by human eyes, and actually, the image can be storedWhen 256 colors with different gray levels are used, if the contrast of an image is higher or lower, the thresholded image becomes a completely black or completely white image, an adaptive threshold method is utilized, the image is traversed and equally divided into a plurality of small rectangles, a plurality of pixels exist in each small rectangle, then the histogram of each pixel is calculated, finally threshold calculation is carried out according to the peak value of the histogram, and the thresholded image is obtained by an interpolation method.
As shown in fig. 3 and 4, the image is subjected to adaptive threshold processing to obtain an image with only two colors, namely, black and white (BGR ═ 0,0,0 and BGR ═ 255,255), and the density of the cars in the parking lot is large and the distance between the cars is small as seen from the original image. Meanwhile, after adaptive binarization processing, the background color and the target color of the image are white and black, and the image is further processed by adopting an expansion function: firstly, setting a 3X 3 matrix, then traversing the image to obtain the value of each pixel, if the pixel is white, the value is marked as 1, if the pixel is black, the value is marked as 0, and then according to the formula
And the 3 x 3 matrix performs an and operation on each pixel, then the black area represented by the target car will decrease and the white background area will increase, and separation between the target cars can be achieved by the dilation process; after the image is processed by the expansion function, black noise still exists, and the black noise is processed by using closed operation.
As shown in fig. 5 and 6, after the image is processed by the close operation, the black noise in the image is basically removed, and at this time, the image is subjected to the negation operation: p (x) 255-h (x), where the two variables are 0 or 255, respectively. In order to further improve the calculation accuracy, the inverted image is processed again by using the expansion function, and the white connected component at this time can represent the automobile target.
As shown in fig. 7 and 8: and after the white connected domain is determined, setting a container, traversing the image, storing all image points in the secondary expansion image in the container, then constructing a four-dimensional container structure, and connecting a line by using points in the four-dimensional container structure as a starting point and an end point. After the containers are arranged and the points in the connected domains are connected, the white connected domains are filled with colors randomly for convenient observation, and finally, the corresponding number of the automobiles is obtained according to the number of the connected domains.
Claims (5)
1. An aerial photography automobile counting method based on OpenCv is characterized in that: the method comprises the following steps:
step 1), shooting a parking lot area by using an unmanned aerial vehicle to obtain an automobile image;
step 2), adjusting the contrast of the automobile image obtained in the step 1), performing Gaussian smoothing on the image, processing the image by using a thermal imaging model, performing gray value transformation on the obtained image, and performing self-adaptive binarization processing on the gray image;
step 3), performing morphological operation on the image obtained in the step 2), processing the image by using a binary function through an expansion function, and processing the image by using a closed operation;
step 4), operating the picture obtained in the step 3) by using an inverse function, and processing the image by using an expansion function;
and 5) carrying out contour extraction on the image obtained in the step 4), and calculating connected domains so as to obtain the number of the automobiles.
2. The OpenCv-based aerial vehicle counting method of claim 1, wherein: in the step 1), the height of the unmanned aerial vehicle is kept between 40 meters and 80 meters in the process of aerial photography of the unmanned aerial vehicle; meanwhile, the included angle between the shooting angle of the camera of the unmanned aerial vehicle and the vertical direction is kept within the range of 20 degrees.
3. The OpenCv-based aerial vehicle counting method of claim 1, wherein: and 3) processing the image by using convolution operation in the expansion function in the step 3), wherein the image is adjusted by the self-adaptive binarization function, three channels B, G, R of the background image are 255, and three channel values of the target image are 0.
4. The OpenCv-based aerial vehicle counting method of claim 1, wherein: after the black area is eliminated in the step 4), the small black spots still exist in the image, and in order to further eliminate the small black spots, the whole image is subjected to negation operation, at this time, black is taken as a background, white is taken as an automobile target, and the white automobile target is further expanded by further utilizing an expansion function.
5. The OpenCv-based aerial vehicle counting method of claim 1, wherein: and 5) setting a container, traversing all image points of the whole image, finding a white connected region, marking the connected region, adding random colors for distinguishing, and finally displaying the number of the color regions in the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110560068.5A CN113269745A (en) | 2021-05-21 | 2021-05-21 | Aerial photography automobile counting method based on OpenCv |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110560068.5A CN113269745A (en) | 2021-05-21 | 2021-05-21 | Aerial photography automobile counting method based on OpenCv |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113269745A true CN113269745A (en) | 2021-08-17 |
Family
ID=77232493
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110560068.5A Pending CN113269745A (en) | 2021-05-21 | 2021-05-21 | Aerial photography automobile counting method based on OpenCv |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113269745A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114926467A (en) * | 2022-07-22 | 2022-08-19 | 新恒汇电子股份有限公司 | Full-automatic lead frame counting method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105894826A (en) * | 2016-06-28 | 2016-08-24 | 侯舸 | Traffic flow density identifying device based on infrared thermal imaging |
| CN105893962A (en) * | 2016-03-31 | 2016-08-24 | 成都信息工程大学 | Method for counting passenger flow at airport security check counter |
| CN106875413A (en) * | 2017-02-13 | 2017-06-20 | 华东师范大学 | A kind of adhesion red blood cell automatic counting method based on high light spectrum image-forming |
| CN107609543A (en) * | 2017-10-24 | 2018-01-19 | 宁夏农林科学院农业经济与信息技术研究所 | A kind of matrimony vine diseases and pests polypide robot scaler |
-
2021
- 2021-05-21 CN CN202110560068.5A patent/CN113269745A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105893962A (en) * | 2016-03-31 | 2016-08-24 | 成都信息工程大学 | Method for counting passenger flow at airport security check counter |
| CN105894826A (en) * | 2016-06-28 | 2016-08-24 | 侯舸 | Traffic flow density identifying device based on infrared thermal imaging |
| CN106875413A (en) * | 2017-02-13 | 2017-06-20 | 华东师范大学 | A kind of adhesion red blood cell automatic counting method based on high light spectrum image-forming |
| CN107609543A (en) * | 2017-10-24 | 2018-01-19 | 宁夏农林科学院农业经济与信息技术研究所 | A kind of matrimony vine diseases and pests polypide robot scaler |
Non-Patent Citations (4)
| Title |
|---|
| 杨鸿捷: "伪彩色图像增强技术在DR胸片影像诊断应用研究", 《中国优秀博硕士学位论文全文数据库(硕士)医药卫生科技辑》 * |
| 盛业华: "《数学形态学理论与地图扫描识别技术》", 30 June 1999, 中国矿业大学出版社 * |
| 陈博: "基于无人机图像的麦穗识别技术研究", 《中国优秀博硕士学位论文全文数据库(硕士)农业科技辑》 * |
| 陈博: "基于阴影的高分辨率遥感图像建筑物提取研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114926467A (en) * | 2022-07-22 | 2022-08-19 | 新恒汇电子股份有限公司 | Full-automatic lead frame counting method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111428748B (en) | HOG feature and SVM-based infrared image insulator identification detection method | |
| CN107507167B (en) | Cargo tray detection method and system based on point cloud plane contour matching | |
| CN110866924B (en) | Line structured light center line extraction method and storage medium | |
| CN111914698B (en) | Human body segmentation method, segmentation system, electronic equipment and storage medium in image | |
| CN107784669A (en) | A kind of method that hot spot extraction and its barycenter determine | |
| CN111415363A (en) | Image edge identification method | |
| CN113761999A (en) | Target detection method and device, electronic equipment and storage medium | |
| CN112529827A (en) | Training method and device for remote sensing image fusion model | |
| CN111881801B (en) | Newly-added construction land remote sensing monitoring method and equipment based on invariant detection strategy | |
| CN101286236B (en) | Infrared object tracking method based on multi- characteristic image and average drifting | |
| CN112446871A (en) | Tunnel crack identification method based on deep learning and OpenCV | |
| CN111027538A (en) | Container detection method based on instance segmentation model | |
| CN113034497A (en) | Vision-based thermos cup weld positioning detection method and system | |
| CN116310095A (en) | Multi-view three-dimensional reconstruction method based on deep learning | |
| CN111507340A (en) | Target point cloud data extraction method based on three-dimensional point cloud data | |
| CN113269745A (en) | Aerial photography automobile counting method based on OpenCv | |
| Wang et al. | Vehicle license plate recognition based on wavelet transform and vertical edge matching | |
| CN113435266A (en) | FCOS intelligent target detection method based on extreme point feature enhancement | |
| Shi et al. | License plate localization in complex environments based on improved GrabCut algorithm | |
| CN116994236A (en) | Low-quality image license plate detection method based on deep neural network | |
| US11699303B2 (en) | System and method of acquiring coordinates of pupil center point | |
| CN113343819B (en) | Efficient unmanned airborne SAR image target segmentation method | |
| CN114862889A (en) | Road edge extraction method and device based on remote sensing image | |
| Rui | Lane line detection technology based on machine vision | |
| CN113052118A (en) | Method, system, device, processor and storage medium for realizing scene change video analysis and detection based on high-speed dome camera |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210817 |