CN105335981B - A kind of cargo monitoring method based on image - Google Patents
A kind of cargo monitoring method based on image Download PDFInfo
- Publication number
- CN105335981B CN105335981B CN201510731904.6A CN201510731904A CN105335981B CN 105335981 B CN105335981 B CN 105335981B CN 201510731904 A CN201510731904 A CN 201510731904A CN 105335981 B CN105335981 B CN 105335981B
- Authority
- CN
- China
- Prior art keywords
- image
- current
- color difference
- abnormal
- monitoring
- 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.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 22
- 230000002159 abnormal effect Effects 0.000 claims abstract description 53
- 238000001914 filtration Methods 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 230000004075 alteration Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 11
- GNFTZDOKVXKIBK-UHFFFAOYSA-N 3-(2-methoxyethoxy)benzohydrazide Chemical compound COCCOC1=CC=CC(C(=O)NN)=C1 GNFTZDOKVXKIBK-UHFFFAOYSA-N 0.000 claims description 9
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000005856 abnormality Effects 0.000 claims description 6
- 238000009825 accumulation Methods 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 230000010339 dilation Effects 0.000 claims description 2
- 238000009499 grossing Methods 0.000 claims description 2
- 239000011159 matrix material Substances 0.000 claims description 2
- 238000005286 illumination Methods 0.000 abstract description 7
- 230000005540 biological transmission Effects 0.000 abstract description 3
- 238000000605 extraction Methods 0.000 abstract description 2
- 238000013467 fragmentation Methods 0.000 abstract 2
- 238000006062 fragmentation reaction Methods 0.000 abstract 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- 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
- G06T2207/30232—Surveillance
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Closed-Circuit Television Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to image procossing and monitoring fields, and in particular to a kind of method that quantity of goods situation of change is monitored using image.The present invention calculates the color difference image I of present image I, current reference image G using CIEDE2000 colour difference formulassub, then to IsubEnhanced, the extraction of binaryzation, filtering process and connected region, then the connected region c to extractingiAbnormal rate R calculating is carried out, is finally compared and analyzed and judged with fragmentation threshold using abnormal rate R, reach the purpose of intelligent alarm.Present invention reduces transmission costs and carrying cost；Since the acquisition of image does not have continuity, therefore even if quantity of goods does not change, the variation of illumination can also make the image of different periods generate larger difference, however the present invention is by introducing CIEDE2000 colour difference formulas, the binary-state threshold with adaptivity, fragmentation threshold and carrying out abnormal tracing detection, influence of the illumination variation to computer judgment can be overcome, improve reliability and adaptability that computer is alerted by picture control.
Description
Technical Field
The invention belongs to the field of image processing and monitoring, and particularly relates to a method for monitoring the change condition of the quantity of goods by using images.
Background
In the existing monitoring technology, a computer intelligently analyzes a video sequence acquired from a camera, so that the understanding of the content in a monitored scene is completed, and the purpose of intelligent warning is achieved. However, transmitting video and storing video adds cost, while for some applications real-time video is not necessary and video acquisition is expensive and difficult. Meanwhile, although the intelligent analysis technology can effectively improve the monitoring efficiency, due to the limitation of the algorithm and the complexity of the field situation, especially the influence of illumination change, the judgment of the equipment on the image is often deviated, and the phenomena of misinformation and missing report are caused. Therefore, the adaptability and reliability of the intelligent analysis algorithm must be improved, and the real situation of the scene can be relatively accurately reflected.
Disclosure of Invention
Compared with video monitoring, the method can greatly reduce data storage capacity, transmission quantity and calculated quantity, quantify abnormal conditions of the goods, and realize unattended monitoring and automatic alarm; and the false alarm phenomenon caused by illumination change can be reduced, and the reliability and the adaptability of monitoring alarm are improved.
In order to achieve the purpose of the invention, the technical scheme adopted by the invention is as follows: an image-based cargo monitoring method, comprising the steps of:
a. the computer obtains the monitoring image I through the cameraoSimultaneously retrieving a reference image Go(ii) a Respectively pairing the monitored images I according to the designated monitored areasoAnd a reference image GoIntercepting to obtain a current image I and a current reference image G; respectively converting the current image I and the current reference image G from RGB color space to CIELAB color space, and calculating the color difference image I of the current image I and the current reference image G by using CIEDE2000 color difference formulasub。
b. For color difference image IsubEnhanced color difference image Ienh(ii) a A binarization threshold value threshold to the chromatic aberration image I is determined according to the mean value mean of the chromatic aberration image, the zero pixel proportion zeroPixelratio of the chromatic aberration image and the non-zero pixel proportion nonZeroPixelratio of the enhanced chromatic aberration imagesubBinarization is carried out to obtain a binarization color difference image I with a foreground and a backgroundbin。
c. For binary color difference image IbinMorphological filtering is performed to eliminate noise interference, and an image o is obtained.
d. Extracting a connected region of the image o by adopting a two-pass scanning method, and recording the extracted connected region as ci,i≥0。
e. To the connected region ciAnd calculating the abnormal rate, wherein the abnormal rate R is the ratio of the accumulated difference area of the images to the total area of the monitored area, and the abnormal rate R is calculated according to the following formula:
wherein N isciIs a connected region ciThe total number of pixel points; i is a current image; n is a radical ofIThe total number of the pixel points of the current image I.
f. Calculating a dynamic threshold, wherein the dynamic threshold is a segmentation threshold and comprises a minimum abnormal rate RminAnd maximum rate of abnormality RmaxThe calculation formula is as follows:
wherein N isobjArea representing the monitored target: segmenting a foreground-monitoring target from the current image I by using a GrabCT algorithm, and then counting the number of pixel points in a connected region in the foreground image, wherein the value is Nobj；NIThe area of the monitored area is the total number of the pixel points of the current image I.
g. Comparing the anomaly rate R with a segmentation threshold respectively:
1) when R is<RminAt the moment, the monitoring area is in a normal state; when R is less than or equal to 0.05 and is more than RminUsing the monitoring image IoReplacing reference image GoThe real-time performance of the reference image is guaranteed, and the influence of accumulation of small changes of the reference image on the reliability of the algorithm is avoided.
2) When R ismin≤R<RmaxAnd when the abnormal area of the monitoring area is higher than the set safety threshold, judging the abnormal state and starting an alarm program.
3) When R is not less than RmaxNamely, the monitored area has a high abnormal rate, which may be caused by a large change of light, or the goods are actually abnormal, and the abnormal state tracking detection is needed to avoid the blind alarm.
Further, in step g, when R is larger than or equal to RmaxThen, the steps of tracking and detecting the abnormal state are as follows:
in step g, when R is not less than RmaxThen, the steps of tracking and detecting the abnormal state are as follows:
g1, obtaining the internal and monitoring chart of the previous N daysLike IoThe image set M without alarm in the same time period has the size S, and S is more than or equal to 0; and setting a loop variable F, and initializing the loop variable F as S.
g2, judging whether F is larger than 0: if F is equal to 0, starting an alarm program; if F > 0, proceed to step g 3.
g3, acquiring the F-1 image in the image set M as the monitoring image IoReference image G ofoThe abnormality rate R is calculated and the F value is reduced by 1.
g4, judging whether the abnormal rate R is larger than the minimum abnormal rate RminIf R > RminShowing the monitored image IoIf the contrast of the image with the F-1 image is abnormal, returning to the step g 2; if R is less than or equal to RminShowing the monitored image IoAnd comparing the image with the F-1 image, wherein no abnormity occurs, and ending the program.
The invention has the following beneficial effects: images in the monitoring area are intercepted and processed, irrelevant information is removed, and the calculation amount can be greatly reduced; calculating the difference between the current image I and the reference image G through a CIEDE2000 color difference formula, and redefining a color difference calculation method through the CIEDE2000 color difference formula, so that the color difference calculation value is closer to the evaluation of human eyes in the whole CIELAB color space; the target contour can be highlighted by carrying out binarization on the color difference image, the binarization threshold values determined by the mean value mean of the color difference image, the zero pixel proportion zeroPixel ratio of the color difference image and the non-zero pixel proportion nonZeroPixelratio of the enhanced color difference image have good adaptivity, the image information can be saved as far as possible for the color difference images with different characteristics, and the effects of reducing the background and noise interference as far as possible can be achieved. By adopting morphological filtering, noise blocks around the target and noise holes in the target in the binary image can be eliminated, and accurate extraction of a connected region is facilitated; quantifying the abnormal state by using the extracted connected region and an abnormal rate calculation formula, and performing comparative analysis by using a dynamic threshold and an abnormal rate R, wherein the dynamic threshold can be set according to the size of a monitoring target; the dynamic threshold is a segmented threshold, so that the abnormal state is refined, and the accuracy and the reliability of the alarm are improved; by passingThe reference image is replaced in real time to avoid the influence of the accumulation of small changes of the reference image on the reliability of the algorithm; when a large-range abnormal condition occurs, the image I is obtained and monitored within the previous N daysoAnd the image set M which is not alarmed at the same time interval carries out abnormal state tracking detection, so that the abnormal rate is calculated under the similar illumination condition, and the false alarm caused by the light change is reduced.
Drawings
FIG. 1 is a block flow diagram of the present invention.
Detailed Description
An image-based cargo monitoring method, as shown in fig. 1, includes the following steps:
a. the computer obtains the monitoring image I through the cameraoSimultaneously retrieving a reference image Go(ii) a Respectively pairing the monitored images I according to the designated monitored areasoAnd a reference image GoIntercepting to obtain a current image I and a current reference image G; respectively converting the current image I and the current reference image G from RGB color space to CIELAB color space, and calculating the color difference image I of the current image I and the current reference image G by using CIEDE2000 color difference formulasub；
b. For color difference image IsubEnhanced color difference image IenhThe method comprises the following specific steps: using a normalized filter with a filter coefficient of K to match the chrominance image IsubSmoothing is carried out, and an enhanced color difference image I is obtained after the smoothed image is squaredenhWherein the filter coefficientsSize of all-1 matrix and color difference image IsubAnd obtaining an enhanced color difference image IenhThe calculation formula of (2):
a binarization threshold value threshold to the chromatic aberration image I is determined according to the mean value mean of the chromatic aberration image, the zero pixel proportion zeroPixelratio of the chromatic aberration image and the non-zero pixel proportion nonZeroPixelratio of the enhanced chromatic aberration imagesubBinarization is carried out to obtain a binarization color difference image I with a foreground and a backgroundbin(ii) a The calculation formulas of the color difference image mean, the color difference image zero pixel proportion zeroPixelratio and the enhanced color difference image non-zero pixel proportion nonZeroPixelratio are respectively as follows:
wherein rows represents the number of rows and cols represents the number of columns;
the binarization threshold value threshold is selected according to the following table:
after the value of the binarization threshold value threshold is determined, the color difference image I is binarizedbinCan be calculated according to the following formula:
c. for binary color difference image IbinMorphological filtering is carried out to eliminate noise interference to obtain an image o, and the method specifically comprises the following steps: using square structural element B of 3x3 first1For binary color difference image IbinTwo etches were performed, followed by a 3 × 3 square structuring element B2Two dilations were performed and the image o after morphological filtering was calculated as follows:
d. extracting a connected region of the image o by adopting a two-pass scanning method, and recording the extracted connected region as ciI is not less than 0, image I is binary color differencebinThe method has the following specific steps that the image o has the foreground and the background, and the image o is scanned twice by adopting a contour marking algorithm:
d1, scanning the pixels in the image o line by line according to the line sequence, judging whether the current pixel is a background, if not, reading a 4-neighborhood pixel set of the current pixel position, judging whether the pixels of the 4 neighborhoods are all marked, if so, designating the label of the current pixel as the minimum label in the 4-neighborhood pixel set; if not, assigning a new label for the current pixel point, and adding 1 to the current maximum label value of the new label; traversing all the pixel points, and finishing the first scanning;
d2, scanning the pixel points in the image o line by line according to the line sequence, judging whether the current pixel point is marked, if so, finding a connected region corresponding to the current label, and assigning the minimum label in the labels belonging to the connected region to the current region; traversing all the pixel points, and finishing the second scanning;
d3, filtering out the connected region smaller than 10 pixels in the step d2 to obtain a connected region ci,i≥0。
e. To the connected region ciCalculating the abnormal rate R which is the total surface of the image accumulated difference area and the monitored areaThe product ratio, the anomaly rate R, is calculated according to the following equation:
wherein,is a connected region ciThe total number of pixel points; i is a current image; n is a radical ofIThe total number of pixel points of the current image I is obtained;
f. calculating a dynamic threshold, wherein the dynamic threshold is a segmentation threshold and comprises a minimum abnormal rate RminAnd maximum rate of abnormality RmaxThe calculation formula is as follows:
wherein N isobjArea representing the monitored target: segmenting a foreground-monitoring target from the current image I by using a GrabCT algorithm, and then counting the number of pixel points in a connected region in the foreground image, wherein the value is Nobj；NIThe area of the monitored area is the total number of the pixel points of the current image I.
g. Comparing the anomaly rate R with a segmentation threshold respectively:
1) when R is<RminAt the moment, the monitoring area is in a normal state; when R is less than or equal to 0.05 and is more than RminUsing the monitoring image IoReplacing reference image GoThe real-time performance of the reference image is ensured, and the influence of the accumulation of small changes of the reference image on the reliability of the algorithm is avoided;
2) when R ismin≤R<RmaxAnd the abnormal area of the monitoring area is higher than the set safety threshold, the monitoring area is judged to be in an abnormal state, an alarm program is started, and alarm information can be sent to a user in a short message mode and the like.
3) When R is not less than RmaxNamely, the monitored area has a high abnormal rate, which may be caused by a large change of light, or the goods are actually abnormal, and the abnormal state tracking detection is needed to avoid the blind alarm.
Preferably, in step g, when R.gtoreq.RmaxThen, the steps of tracking and detecting the abnormal state are as follows:
g1, obtaining the image I within N days before and monitoringoThe image set M without alarm in the same time period has the size S, and S is more than or equal to 0; and setting a loop variable F, and initializing the loop variable F as S.
g2, judging whether F is larger than 0: if F is equal to 0, starting an alarm program; if F > 0, proceed to step g 3. If the initial value of F is 0, the obtained image set M is indicated to be an empty set, and at the moment, an alarm program can be started to inform the staff of monitoring the image IoAnd (4) carrying out manual judgment to prevent missing report.
g3, acquiring the F-1 image in the image set M as the monitoring image IoReference image G ofoThe abnormality rate R is calculated and the F value is reduced by 1.
g4, judging whether the abnormal rate R is larger than the minimum abnormal rate RminIf R > RminShowing the monitored image IoIf the image comparison with the F-1 st image is abnormal, returning to the step g2, since the step g3 decreases the value of F by 1 each time, F is gradually reduced, and if the final value of F is 0, the monitoring image I is indicatedoComparing and analyzing all the images in the image set M, wherein the comparison result is abnormal each time, and finally when the step g2 is returned, an alarm program is started to send an alarm message that the abnormal tracking detection result is abnormal to the user; if R is less than or equal to RminShowing the monitored image IoThe image is compared with the F-1 image without abnormality, which indicates that the monitoring image IoThe large-range exception is caused by light change, and the program is ended to prevent false alarm.
The invention takes the image as the basis for judging the change of the quantity of the goods in the monitoring area, and greatly reduces the transmission cost and the storage cost compared with the video monitoring; because the acquisition of the image is not continuous, even if the number of goods is not changed, the change of illumination can cause the images in different time periods to generate larger difference, however, the invention overcomes the influence of the illumination change on the judgment of the computer by introducing a CIEDE2000 color difference formula, a binarization threshold value with self-adaptability and a segmentation threshold value and carrying out abnormal tracking detection, and improves the reliability and the adaptability of the computer for monitoring and alarming through the image.
Claims (7)
1. An image-based cargo monitoring method, comprising the steps of:
a. the computer obtains the monitoring image I through the cameraoSimultaneously retrieving a reference image Go(ii) a Respectively pairing the monitored images I according to the designated monitored areasoAnd a reference image GoIntercepting to obtain a current image I and a current reference image G; respectively converting the current image I and the current reference image G from RGB color space to CIELAB color space, and calculating the color difference image I of the current image I and the current reference image G by using CIEDE2000 color difference formulasub；
b. For color difference image IsubEnhancing to obtain enhanced color difference image Ienh(ii) a A binarization threshold value threshold to the chromatic aberration image I is determined according to the mean value mean of the chromatic aberration image, the zero pixel proportion zeroPixelratio of the chromatic aberration image and the non-zero pixel proportion nonZeroPixelratio of the enhanced chromatic aberration imagesubBinarization is carried out to obtain a binarization color difference image I with a foreground and a backgroundbin；
c. For binary color difference image IbinPerforming morphological filtering to eliminate noise interference to obtain an image o;
d. extracting a connected region of the image o by adopting a two-pass scanning method, and recording the extracted connected region as ci,i≥0；
e. To the connected region ciAnd calculating the abnormal rate, wherein the abnormal rate R is the ratio of the accumulated difference area of the images to the total area of the monitored area, and the abnormal rate R is calculated according to the following formula:
wherein,is a connected region ciThe total number of pixel points; i is a current image; n is a radical ofIThe total number of pixel points of the current image I is obtained;
f. calculating a dynamic threshold, wherein the dynamic threshold is a segmentation threshold and comprises a minimum abnormal rate RminAnd maximum rate of abnormality RmaxThe calculation formula is as follows:
wherein N isobjArea representing the monitored target: segmenting a foreground-monitoring target from the current image I by using a GrabCT algorithm, and then counting the number of pixel points in a connected region in the foreground image, wherein the value is Nobj；NIThe area of the monitoring area is the total number of pixel points of the current image I;
g. comparing the anomaly rate R with a segmentation threshold respectively:
1) when R is<RminAt the moment, the monitoring area is in a normal state; when R is less than or equal to 0.05 and is more than RminUsing the monitoring image IoReplacing reference image GoThe real-time performance of the reference image is ensured, and the influence of the accumulation of small changes of the reference image on the reliability of the algorithm is avoided;
2) when R ismin≤R<RmaxIf the abnormal area of the monitoring area is higher than the set safety threshold, judging the monitoring area to be in an abnormal state, and starting an alarm program;
3) when R is not less than RmaxNamely, the monitored area has a high abnormal rate, which may be caused by a large change of light, or the goods are actually abnormal, and the abnormal state tracking detection is needed to avoid the blind alarm.
2. The image-based cargo monitoring method of claim 1, wherein:
in step g, when R is not less than RmaxThen, the steps of tracking and detecting the abnormal state are as follows:
g1, obtaining the image I within N days before and monitoringoThe image set M without alarm in the same time period has the size S, and S is more than or equal to 0; setting a loop variable F, and initializing the loop variable F as S;
g2, judging whether F is larger than 0: if F is equal to 0, starting an alarm program; if F > 0, go to step g 3;
g3, acquiring the F-1 image in the image set M as the monitoring image IoReference image G ofoCalculating an abnormal rate R, and subtracting 1 from the value F;
g4, judging whether the abnormal rate R is larger than the minimum abnormal rate RminIf R > RminShowing the monitored image IoIf the contrast of the image with the F-1 image is abnormal, returning to the step g 2; if R is less than or equal to RminShowing the monitored image IoAnd comparing the image with the F-1 image, wherein no abnormity occurs, and ending the program.
3. The method according to claim 1The image cargo monitoring method is characterized in that: in the step b, a normalized filter with the filter coefficient of K is adopted to match the chromatic aberration image IsubSmoothing is carried out, and an enhanced color difference image I is obtained after the smoothed image is squaredenhWherein the filter coefficientsSize of all-1 matrix and color difference image IsubAnd (4) according to the obtained result, obtaining an enhanced color difference image IenhThe calculation formula of (2):
4. the image-based cargo monitoring method of claim 1, wherein: in the step b, the calculation formulas of the color difference image mean, the color difference image zero pixel ratio zeropixel ratio and the enhanced color difference image non-zero pixel ratio nonzeropixel ratio are respectively as follows:
where rows represents the number of rows and cols represents the number of columns.
5. The image-based cargo monitoring method of claim 1, wherein: in the step b, the binary threshold value threshold is valued according to the following table:
after the value of the binarization threshold value threshold is determined, the color difference image I is binarizedbinCan be calculated according to the following formula:
6. the image-based cargo monitoring method of claim 1, wherein: step c is to use square structural element B of 3x31For binary color difference image IbinTwo etches were performed, followed by a 3 × 3 square structuring element B2Two dilations were performed and the image o after morphological filtering was calculated as follows: o (((I))bin⊕B1)⊕B1)ΘB2)ΘB2。
7. The image-based cargo monitoring method of claim 1, wherein: the image o in the step d has a foreground and a background, and the image o is scanned twice in sequence by adopting a contour marking algorithm, which specifically comprises the following steps:
d1, scanning the pixels in the image o line by line according to the line sequence, judging whether the current pixel is a background, if not, reading a 4-neighborhood pixel set of the current pixel position, judging whether the pixels of the 4 neighborhoods are all marked, if so, designating the label of the current pixel as the minimum label in the 4-neighborhood pixel set; if not, assigning a new label for the current pixel point, and adding 1 to the current maximum label value of the new label; traversing all the pixel points, and finishing the first scanning;
d2, scanning the pixel points in the image o line by line according to the line sequence, judging whether the current pixel point is marked, if so, finding a connected region corresponding to the current label, and assigning the minimum label in the labels belonging to the connected region to the current region; traversing all the pixel points, and finishing the second scanning;
d3, filtering out the connected region smaller than 10 pixels in the step d2 to obtain a connected region ci,i≥0。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510731904.6A CN105335981B (en) | 2015-10-29 | 2015-10-29 | A kind of cargo monitoring method based on image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510731904.6A CN105335981B (en) | 2015-10-29 | 2015-10-29 | A kind of cargo monitoring method based on image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105335981A CN105335981A (en) | 2016-02-17 |
| CN105335981B true CN105335981B (en) | 2018-06-29 |
Family
ID=55286486
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510731904.6A Active CN105335981B (en) | 2015-10-29 | 2015-10-29 | A kind of cargo monitoring method based on image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105335981B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106226249A (en) * | 2016-09-29 | 2016-12-14 | 成都齐力丝绸有限公司 | A kind of color digitized Compare System and method |
| CN107464080A (en) * | 2017-08-14 | 2017-12-12 | 四川鹏尧智运科技有限公司 | High in the clouds supervisory systems available for ships that transport |
| CN108038989A (en) * | 2017-12-04 | 2018-05-15 | 杭州纳戒科技有限公司 | Shared logistics box control method, apparatus and system |
| CN108053398A (en) * | 2017-12-19 | 2018-05-18 | 南京信息工程大学 | A kind of melanoma automatic testing method of semi-supervised feature learning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101123722B (en) * | 2007-09-25 | 2010-12-01 | 北京智安邦科技有限公司 | Panorama video intelligent monitoring method and system |
| CN201717987U (en) * | 2009-10-14 | 2011-01-19 | 深圳辉锐天眼科技有限公司 | Villa intelligent monitoring system all-in-one machine |
| WO2013141067A1 (en) * | 2012-03-23 | 2013-09-26 | コニカミノルタ株式会社 | Image-generating apparatus |
-
2015
- 2015-10-29 CN CN201510731904.6A patent/CN105335981B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105335981A (en) | 2016-02-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Park et al. | Single image dehazing with image entropy and information fidelity | |
| CN113808138B (en) | Artificial intelligence-based wire and cable surface defect detection method | |
| CN104966304B (en) | Multi-target detection tracking based on Kalman filtering and nonparametric background model | |
| US20210392261A1 (en) | Flicker mitigation via image signal processing | |
| CN109308447A (en) | The method of equipment operating parameter and operating status is automatically extracted in remote monitoriong of electric power | |
| US20090027521A1 (en) | Image processing apparatus, image-taking system, image processing method and image processing program | |
| CN105335981B (en) | A kind of cargo monitoring method based on image | |
| CN107767390B (en) | The shadow detection method and its system of monitor video image, shadow removal method | |
| CN106682665B (en) | Seven-segment type digital display instrument number identification method based on computer vision | |
| CN111402247A (en) | Machine vision-based method for detecting defects of suspension clamp on power transmission line | |
| CN107085714A (en) | A kind of forest fire detection method based on video | |
| CN107742307A (en) | Based on the transmission line galloping feature extraction and parameters analysis method for improving frame difference method | |
| CN116977342B (en) | PCB circuit detection method based on image segmentation | |
| US8482630B2 (en) | Apparatus and method for adjusting automatic white balance by detecting effective area | |
| JP2010200295A (en) | Method and device for maintaining image background by using multiple gaussian distribution models | |
| CN110188640B (en) | Face recognition method, face recognition device, server and computer readable medium | |
| CN113936252A (en) | Battery car intelligent management system and method based on video monitoring | |
| CN113744326B (en) | Fire detection method based on seed region growth rule in YCRCB color space | |
| CN104299234B (en) | The method and system that rain field removes in video data | |
| WO2024016632A1 (en) | Bright spot location method, bright spot location apparatus, electronic device and storage medium | |
| CN112037137A (en) | Method and device for eliminating fuzzy region of insulator disc surface edge in infrared image | |
| CN111145219B (en) | Efficient video moving target detection method based on Codebook principle | |
| KR101521269B1 (en) | Method for detecting snow or rain on video | |
| CN112926676B (en) | False target identification method and device and computer equipment | |
| CN102982558A (en) | Method and device for detecting moving target |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: No. 51 Dapingzheng Street, Yuzhong District, Chongqing, 400000 Patentee after: Zhongdian Zhi'an Technology Co.,Ltd. Address before: No. 51 Dapingzheng Street, Yuzhong District, Chongqing, 400000 Patentee before: CHONGQING TELECOMMUNICATION SYSTEM INTEGRATION CO.,LTD. |
|
| CP01 | Change in the name or title of a patent holder |