CN109272489A - Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background - Google Patents
Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background Download PDFInfo
- Publication number
- CN109272489A CN109272489A CN201810956335.9A CN201810956335A CN109272489A CN 109272489 A CN109272489 A CN 109272489A CN 201810956335 A CN201810956335 A CN 201810956335A CN 109272489 A CN109272489 A CN 109272489A
- Authority
- CN
- China
- Prior art keywords
- image
- local entropy
- background
- infrared
- notable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- 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
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20164—Salient point detection; Corner detection
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention proposes a kind of method for detecting infrared puniness target inhibited based on background with multiple dimensioned local entropy, and infrared image is normalized first;Secondly guiding figure filter filtering is carried out to infrared image, the image after background inhibits is obtained after difference;Then by calculating multiple dimensioned local entropy weight figure, the maximum value of different scale local entropy notable figure is taken to same location of pixels, obtains final local entropy weight figure;Image after inhibiting later to background is multiplied with local entropy notable figure, obtains the notable figure of infrared small target;Finally notable figure is filtered using Susan filter, removes isolated bright spot, the non-zero region of image is Small object region after processing.The present invention can be used for detecting the Weak target in infrared image, can effectively improve the detection accuracy of Infrared Images target.
Description
Technical field
The invention belongs to technical field of image processing, further relate under complex background in infrared image processing field
A kind of method for detecting infrared puniness target inhibited based on background with multiple dimensioned local entropy.
Background technique
Infrared small target detection has been widely used in multiple fields, such as infrared object tracking, precise guidance and long-range morning
Phase early warning etc..Due to the complexity of imaging circumstances and the characteristic of infrared picture itself, infrared small target stock size is small, signal
It is weak, background is complicated so that its detection and tracking is extremely difficult.Therefore, small IR targets detection is always infrared image processing
The hot and difficult issue of area research.
Research in terms of human eye vision attention mechanism in recent years, is widely used in the detection of infrared small target.
Visual attention model is a kind of biological heuristic models proposed earliest by Koch etc. in 1987.Vision noticing mechanism is a mankind
To the important psychological regulation mechanism that most interested, most relevant range is selected and paid attention in scene.In 1998, Itti
Et al. first proposed a kind of visual attention model, and model is applied to the image detection of visible light, receives the wide of people
General concern.The it is proposeds such as Itti are a kind of conspicuousness detection algorithms based on center-surround, and the main thought of algorithm is
By the brightness of image, color and three, direction feature information extraction, and multiple dimensioned gaussian pyramid is formed, to obtain significant
Figure carries out target detection.Cheng in 2011 etc. proposes a kind of conspicuousness detection algorithm based on global contrast, mentions in algorithm
Spatial relationship is also extracted as a kind of feature, causes extensive concern by the concept of local contrast out.Later
Chen etc. proposes a kind of local contrast algorithm (LCM), and vision significance is shown up prominently in terms of Dim targets detection.Algorithm is logical
It crosses central area and 8 neighborhood of surrounding compares to obtain local contrast figure, then carry out target detection again, but vulnerable to influence of noise, it is empty
Alert rate is higher.Han etc. proposes a kind of improved LCM (ILCM) algorithm, algorithm is first by image procossing on the basis of LCM algorithm
The maximum value in LCM algorithm is replaced at sub-pixel block, then with mean value, so that false alarm rate decreases, but still vulnerable to noise shadow
It rings.
Summary of the invention
For the above the deficiencies in the prior art, the present invention proposes a kind of to inhibit infrared with multiple dimensioned local entropy based on background
Detection method of small target, to improve detection performance.The present invention can be used for detecting the Weak target in infrared image, energy
The effective detection accuracy for improving Infrared Images target.
Realizing technical thought of the invention is: firstly, infrared image is normalized;Secondly, to infrared image
Guiding figure filter filtering is carried out, the image after background inhibits is obtained after difference;Then, by calculating multiple dimensioned local entropy weight
Value figure, the maximum value of different scale local entropy notable figure is taken to same location of pixels, obtains final local entropy weight figure;It
Afterwards, the image after inhibiting to background is multiplied with local entropy notable figure, obtains the notable figure of infrared small target;Finally, utilizing
Susan filter is filtered notable figure, removes isolated bright spot, and the non-zero region of image is Small object area after processing
Domain.
The technical solution of the present invention is as follows:
A kind of method for detecting infrared puniness target inhibited based on background with multiple dimensioned local entropy, it is characterised in that:
The following steps are included:
Step 1: one infrared image I to be detected of input;
Step 2: the image I of input being converted into gray level image and is normalized, image I is obtainedin;
Step 3: background inhibition is carried out by following steps:
Step 3.1: using guiding figure filter, to the infrared image I after normalizationinIt is filtered, after obtaining filtering
Class background image IB;
Step 3.2: by the infrared image I after normalizationinWith the class background image I obtained after filteringBDifference is carried out, is obtained
Image I after inhibiting to backgroundBS;
Step 4: multiple dimensioned local entropy is calculated by following steps:
Step 4.1: selected window scale is the window of m to the infrared image I after normalizationinCarry out local entropy operation;
Step 4.2: replacing the value of window center pixel with the local entropy being calculated, obtain the part that window size is m
Entropy weight figure Wm;
Step 4.3: changing the value of window size m, repeat step 4.1 and step 4.2, obtain under different windows scale
Local entropy weight figure;
Step 4.4: for a certain location of pixels of local entropy weight figure, taking in all local entropy weight figures in the pixel position
The maximum value for setting place forms final local entropy weight figure ILE;
Step 5: the image I after inhibiting to backgroundBSWith local entropy notable figure ILEIt is multiplied, obtains infrared small target
Notable figure S;
Step 6: target detection is carried out by following procedure:
Step 6.1: notable figure S being filtered using Susan filter to obtain image Iout;
Step 6.2: removal image IoutIn isolated noise point after, obtain image IoutIn non-zero region be Small object
Region.
Further preferred embodiment, a kind of small IR targets detection inhibited based on background with multiple dimensioned local entropy
Method, it is characterised in that: using guiding figure filter in step 3.1, to the infrared image I after normalizationinIt is filtered, obtains
Class background image I after filteringBDetailed process are as follows:
With guiding figure filter to the infrared image I after normalizationinIt is filtered, obtains background image:
Wherein, l is integer, and w indicates window size, FwIndicate that guiding figure filter when window size is w, * indicate filter
Wave operation;
Then change window size and repeat filtering operation, obtain multiple background images;Finally make even to multiple background images
Mean value obtains final class background image IB:
。
Further preferred embodiment, a kind of small IR targets detection inhibited based on background with multiple dimensioned local entropy
Method, it is characterised in that: step 4.1 to 4.2 obtains the local entropy weight figure W that window size is mmProcess are as follows:
Selected window scale is the window of m to the infrared image I after normalization firstinFrom top to bottom, it from left to right carries out
Window is mobile: if image IinSize be M × N, by the neighborhood definition of pixel (x, y) are as follows:
θm=(p, q) | max (| p-x |, | q-y |)≤m, m=1,2 ..., L
Wherein m, L are positive integer, and m indicates the scale size of window, and (p, q) indicates picture in the neighborhood of pixel (x, y)
The coordinate position of vegetarian refreshments;
Then the local entropy for calculating pixel (x, y) neighborhood, pixel (x, y) neighborhood local entropy when window size is m
Is defined as:
Wherein D is indicated when window size is m, neighborhood θmIn include D kind gray value, i-th kind of gray value is fi, i=1,
2 ..., D, f (x, y) indicate pixel (x, y) corresponding gray value,piIt is probability density function, niFor i-th kind of ash
The number of pixels of angle value.
Further preferred embodiment, a kind of small IR targets detection inhibited based on background with multiple dimensioned local entropy
Method, it is characterised in that: the image I after inhibiting in step 5 to backgroundBSWith local entropy notable figure ILEIt is multiplied, obtains infrared
The detailed process of the notable figure S of Small object are as follows:
The value of notable figure at point (x, y) is indicated first are as follows:
S (x, y)=IBS(x,y)×ILE(x,y)
Wherein ILE(x, y) indicates the multiple dimensioned local entropy weight of pixel (x, y), IBS(x, y) indicates pixel (x, y)
Value after background inhibition, S (x, y) indicate the value of the notable figure of pixel (x, y) infrared small target;
Secondly value minus in notable figure is set to zero:
。
Beneficial effect
The present invention compared with prior art, has the advantage that
1) guiding figure filter is applied in the scene for extracting image background information for the first time, and background is obtained by difference
Infrared target notable figure after inhibition;
2) Small object in image is highlighted using the notable figure that improved local entropy is constituted, and uses multiple dimensioned office
Portion's entropy adapts to the dynamic change of Small object size under the conditions of different background, detection error caused by avoiding window selection improper;
3) final notable figure is to weight, the deficiency of the single treatment mode avoided, while passing through Susan filter
Processing eliminate influence of the background clutter to target detection.
Additional aspect and advantage of the invention will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect of the invention and advantage will become from the description of the embodiment in conjunction with the following figures
Obviously and it is readily appreciated that, in which:
Fig. 1 is the basic procedure of small IR targets detection.
Fig. 2 is the result of small IR targets detection: first is classified as original infrared input picture;Second is classified as by place
The Saliency maps picture obtained after reason;Third is classified as testing result.
Specific embodiment
The embodiment of the present invention is described below in detail, the embodiment is exemplary, it is intended to it is used to explain the present invention, and
It is not considered as limiting the invention.
Referring to Fig.1, one of the present embodiment inhibits the small IR targets detection with multiple dimensioned local entropy based on background
Method the following steps are included:
Step 1: one infrared image I to be detected of input.
Step 2: the image I of input being converted into gray level image and is normalized, image I is obtainedin.I.e. to infrared
Image I carries out gray scale and normalized, and the gray value of infrared image I is normalized between 0-1, image I is obtainedin。
Step 3: background inhibits.
I.e. using guiding figure filter to the infrared image I after normalizationinIt is filtered.Guiding figure filter is one
The relatively new type image filter of kind, it can not only keep the edge feature of image smoothing well, and can also be preferable
Retain the information of edge neighborhood in image.It is different in general filter, and operation efficiency is unrelated with the size of core, thus
Many fields are widely used, such as: image defogging scratches figure, except make an uproar etc..
When being oriented among treatment process of the figure filter applied to infrared image, due to infrared small target in infrared image
Itself the characteristics of, it may be assumed that pixel shared by infrared target is considerably less, and the brightness of infrared small target is relatively high etc..It will be infrared
Image passes through after guiding figure filtering, can preferably save the background information etc. in image, finally, by making with original image
Difference obtains the image after background inhibits.
Step 3.1: using guiding figure filter, to the infrared image I after normalizationinIt is filtered, after obtaining filtering
Class background image IB.Detailed process are as follows:
With guiding figure filter to the infrared image I after normalizationinIt is filtered, obtains background image:
Wherein, l is integer, and w indicates window size, FwIndicate that guiding figure filter when window size is w, * indicate filter
Wave operation.
Then change window size and repeat filtering operation, obtain multiple background images;Finally make even to multiple background images
Mean value obtains final class background image IB:
Step 3.2: by the infrared image I after normalizationinWith the class background image I obtained after filteringBDifference is carried out, is obtained
Image I after inhibiting to backgroundBS;IBS=Iin-IB。
Step 4: multiple dimensioned local entropy calculates:
In one width infrared image, Weak target can regard a singular point incompatible with ambient enviroment, target area as
Target context and background can be different from by the difference of gray value.But when the complicated noise jamming of background is strong, small mesh
Mark is easy to be submerged, and the difference of gray value is not enough to detect Small object at this time, may cause the case where inspection does not measure Small object
Occur.And window local entropy is selected then to avoid the generation of such case.
The information content contained by different size of window is proportional to the numerical value of the entropy of window in image, the letter that window includes
Breath content is influenced by the selection (i.e. window size) of window.In order to adapt to the detection of different size Small objects, weaken clutter
Bring adverse effect, significantly counteracting residual noise.Local entropy operation is carried out using different size of window, is videlicet carried out
Calculate multiple dimensioned local entropy.
Step 4.1: selected window scale is the window of m to the infrared image I after normalizationinCarry out local entropy operation.
Step 4.2: replacing the value of window center pixel with the local entropy being calculated, obtain the part that window size is m
Entropy weight figure Wm;For edge pixel point using corresponding edge pixel filling window come the local entropy of calculating field.
Step 4.1 to 4.2 obtain window size be m local entropy weight figure WmProcess are as follows:
Selected window scale is the window of m to the infrared image I after normalization firstinFrom top to bottom, it from left to right carries out
Window is mobile: if image IinSize be M × N, by the neighborhood definition of pixel (x, y) are as follows:
θm=(p, q) | max (| p-x |, | q-y |)≤m, m=1,2 ..., L
Wherein m, L are positive integer, and m indicates the scale size of window, and (p, q) indicates picture in the neighborhood of pixel (x, y)
The coordinate position of vegetarian refreshments.
Then the local entropy for calculating pixel (x, y) neighborhood, pixel (x, y) neighborhood local entropy when window size is m
Is defined as:
Wherein D is indicated when window size is m, neighborhood θmIn include D kind gray value, i-th kind of gray value is fi, i=1,
2 ..., D, f (x, y) indicate pixel (x, y) corresponding gray value,piIt is probability density function, niFor i-th kind of ash
The number of pixels of angle value.
Step 4.3: changing the value of window size m, repeat step 4.1 and step 4.2, obtain under different windows scale
Local entropy weight figure.
Step 4.4: for a certain location of pixels (x, y) of local entropy weight figure, taking in all local entropy weight figures at this
Maximum value at location of pixels (x, y) forms final local entropy weight figure ILE;
ILE(x, y)=max | Wm(x, y) |, m=1,2 ..., L }
Step 5: the notable figure of infrared small target:
Vision significance measurement is the key link in the image partition method of view-based access control model conspicuousness, mainly calculates image
In each pixel conspicuousness, as a result with indicating with the equal-sized width gray level image of original input picture, referred to as notable figure.
Wherein each pixel value represents the saliency value of corresponding position pixel in original image, is worth bigger explanation pixel in original graph
It is more significant as in, more it is easy to get the attention of observer.
Image I after inhibiting to backgroundBSWith local entropy notable figure ILEIt is multiplied, obtains the notable figure of infrared small target
S.Detailed process are as follows:
The value of notable figure can indicate first at point (x, y) are as follows:
S (x, y)=IBS(x,y)×ILE(x,y)
Wherein ILE(x, y) indicates the multiple dimensioned local entropy weight of pixel (x, y), IBS(x, y) indicates pixel (x, y)
Value after background inhibition, S (x, y) indicate the value of the notable figure of pixel (x, y) infrared small target.
Secondly value minus in notable figure is set to zero:
Then whole image can be traversed by, which handling by multiple dimensioned local entropy, realizes highlighting while inhibiting background to Small object
Noise improves the adaptability of algorithm and the robustness of multiple dimensioned local contrast.
Step 6: target detection:
Susan filter is a kind of detection method based on template window, mainly by each pixel point of image
The place of setting establishes a window, by comparing neighborhood of pixel points with its center similarity degree, to carry out image procossing.Used ginseng
Number is very few, calculation amount and storage capacity is required low.Mainly to be directed to edge detection and Corner Detection, it is also used for making an uproar
Sound is eliminated.The non-maxima suppression of edge detection in Susan filter is removed, is used to by the characteristic based on infrared small object
Carry out target detection.
Target detection is carried out by following procedure:
Step 6.1: notable figure S being filtered using Susan filter to obtain image Iout;
Step 6.2: statistics IoutNon-zero pixels number in image is directly entered step 6.3 if number is 1, no
Then carry out rejecting IoutThe operation of isolated noise point in image, even image IoutIn 8 neighborhoods around certain non-zero pixels point
Pixel value is 0, then it is assumed that the pixel is isolated noise spot, carries out rejecting processing to it, its value is set to 0.
Step 6.3: taking image IoutIn non-zero region be Small object region.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art are not departing from the principle of the present invention and objective
In the case where can make changes, modifications, alterations, and variations to the above described embodiments within the scope of the invention.
Claims (4)
1. a kind of inhibited and the method for detecting infrared puniness target of multiple dimensioned local entropy based on background, it is characterised in that: including with
Lower step:
Step 1: one infrared image I to be detected of input;
Step 2: the image I of input being converted into gray level image and is normalized, image I is obtainedin;
Step 3: background inhibition is carried out by following steps:
Step 3.1: using guiding figure filter, to the infrared image I after normalizationinIt is filtered, the class after being filtered
Background image IB;
Step 3.2: by the infrared image I after normalizationinWith the class background image I obtained after filteringBDifference is carried out, is carried on the back
Image I after scape inhibitionBS;
Step 4: multiple dimensioned local entropy is calculated by following steps:
Step 4.1: selected window scale is the window of m to the infrared image I after normalizationinCarry out local entropy operation;
Step 4.2: replacing the value of window center pixel with the local entropy being calculated, obtain the local entropy weight that window size is m
Value figure Wm;
Step 4.3: changing the value of window size m, repeat step 4.1 and step 4.2, obtain the part under different windows scale
Entropy weight figure;
Step 4.4: for a certain location of pixels of local entropy weight figure, taking in all local entropy weight figures in the pixel position
Maximum value, form final local entropy weight figure ILE;
Step 5: the image I after inhibiting to backgroundBSWith local entropy notable figure ILEIt is multiplied, obtains the notable figure of infrared small target
S;
Step 6: target detection is carried out by following procedure:
Step 6.1: notable figure S being filtered using Susan filter to obtain image Iout;
Step 6.2: removal image IoutIn isolated noise point after, obtain image IoutIn non-zero region be Small object area
Domain.
2. a kind of method for detecting infrared puniness target inhibited based on background with multiple dimensioned local entropy according to claim 1,
It is characterized by: using guiding figure filter in step 3.1, to the infrared image I after normalizationinIt is filtered, is filtered
Class background image I laterBDetailed process are as follows:
With guiding figure filter to the infrared image I after normalizationinIt is filtered, obtains background image:
IBw=Iin*Fw, w=w1,w2,……,wl
Wherein, l is integer, and w indicates window size, FwIndicate that guiding figure filter when window size is w, * indicate filtering fortune
It calculates;
Then change window size and repeat filtering operation, obtain multiple background images;Finally multiple background images are averaged
Obtain final class background image IB:
。
3. a kind of method for detecting infrared puniness target inhibited based on background with multiple dimensioned local entropy according to claim 1,
It is characterized by: step 4.1 to 4.2 obtain window size be m local entropy weight figure WmProcess are as follows:
Selected window scale is the window of m to the infrared image I after normalization firstinFrom top to bottom, window is from left to right carried out
It is mobile: if image IinSize be M × N, by the neighborhood definition of pixel (x, y) are as follows:
θm=(p, q) | max (| p-x |, | q-y |)≤m, m=1,2 ..., L
Wherein m, L are positive integer, and m indicates the scale size of window, and (p, q) indicates pixel in the neighborhood of pixel (x, y)
Coordinate position;
Then the local entropy for calculating pixel (x, y) neighborhood, the definition of pixel (x, y) neighborhood local entropy when window size is m
Are as follows:
Wherein D is indicated when window size is m, neighborhood θmIn include D kind gray value, i-th kind of gray value is fi, i=1,2 ...,
D, f (x, y) indicate pixel (x, y) corresponding gray value,piIt is probability density function, niFor i-th kind of gray value
Number of pixels.
4. a kind of method for detecting infrared puniness target inhibited based on background with multiple dimensioned local entropy according to claim 1,
It is characterized by: the image I after inhibiting in step 5 to backgroundBSWith local entropy notable figure ILEIt is multiplied, obtains infrared small mesh
The detailed process of target notable figure S are as follows:
The value of notable figure at point (x, y) is indicated first are as follows:
S (x, y)=IBS(x,y)×ILE(x,y)
Wherein ILE(x, y) indicates the multiple dimensioned local entropy weight of pixel (x, y), IBS(x, y) indicates the suppression of pixel (x, y) background
Value after system, S (x, y) indicate the value of the notable figure of pixel (x, y) infrared small target;
Secondly value minus in notable figure is set to zero:
。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810956335.9A CN109272489B (en) | 2018-08-21 | 2018-08-21 | Infrared weak and small target detection method based on background suppression and multi-scale local entropy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810956335.9A CN109272489B (en) | 2018-08-21 | 2018-08-21 | Infrared weak and small target detection method based on background suppression and multi-scale local entropy |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109272489A true CN109272489A (en) | 2019-01-25 |
CN109272489B CN109272489B (en) | 2022-03-29 |
Family
ID=65154148
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810956335.9A Active CN109272489B (en) | 2018-08-21 | 2018-08-21 | Infrared weak and small target detection method based on background suppression and multi-scale local entropy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109272489B (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109903272A (en) * | 2019-01-30 | 2019-06-18 | 西安天伟电子系统工程有限公司 | Object detection method, device, equipment, computer equipment and storage medium |
CN109949292A (en) * | 2019-03-20 | 2019-06-28 | 湖南镭目科技有限公司 | A kind of detection method and device of steel strip surface defect |
CN110135312A (en) * | 2019-05-06 | 2019-08-16 | 电子科技大学 | A kind of quick small target detecting method based on classification LCM |
CN110490848A (en) * | 2019-08-02 | 2019-11-22 | 上海海事大学 | Infrared target detection method, apparatus and computer storage medium |
CN110674782A (en) * | 2019-09-30 | 2020-01-10 | 电子科技大学 | Infrared dim target detection method based on fractional entropy |
CN110717499A (en) * | 2019-09-24 | 2020-01-21 | 南京理工大学 | Image feature extraction method based on single-layer perceptron model |
CN111160260A (en) * | 2019-12-30 | 2020-05-15 | 湖北航天技术研究院总体设计所 | SAR image target detection method and system |
CN111260624A (en) * | 2020-01-15 | 2020-06-09 | 陕西师范大学 | Multi-scale infrared small target detection method based on feature matching and scale selection strategy |
CN111369458A (en) * | 2020-02-28 | 2020-07-03 | 中国人民解放军空军工程大学 | Infrared dim target background suppression method based on multi-scale rolling guide filtering smoothing |
CN112163606A (en) * | 2020-09-17 | 2021-01-01 | 哈尔滨工业大学(威海) | Infrared small target detection method based on block contrast weighting |
CN112766109A (en) * | 2021-01-08 | 2021-05-07 | 北京航空航天大学 | Infrared weak and small target self-adaptive detection method |
CN113076888A (en) * | 2021-04-09 | 2021-07-06 | 中国电子科技集团公司第二十八研究所 | Multi-scale local binary model photoelectric image target detection method |
CN113205494A (en) * | 2021-04-27 | 2021-08-03 | 武汉大学 | Infrared small target detection method and system based on adaptive scale image block weighting difference measurement |
CN113643315A (en) * | 2021-07-22 | 2021-11-12 | 武汉大学 | Infrared small target detection method based on self-adaptive peak gradient descent filter |
CN113658191A (en) * | 2021-07-05 | 2021-11-16 | 中国人民解放军火箭军工程大学 | Infrared dim target detection method based on local probability hypergraph dissimilarity measure |
CN113822352A (en) * | 2021-09-15 | 2021-12-21 | 中北大学 | Infrared dim target detection method based on multi-feature fusion |
CN116645580A (en) * | 2023-06-05 | 2023-08-25 | 北京邮电大学 | Method and device for detecting infrared dim and small targets based on space-time characteristic difference |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5384716A (en) * | 1992-09-24 | 1995-01-24 | Matsushita Electric Industrial Co., Ltd. | Occupant condition determining apparatus |
CN102819740A (en) * | 2012-07-18 | 2012-12-12 | 西北工业大学 | Method for detecting and positioning dim targets of single-frame infrared image |
CN103632370A (en) * | 2013-12-02 | 2014-03-12 | 西北工业大学 | Multilevel iteration visualization optimization based total-focusing synthetic aperture perspective imaging method |
CN104268844A (en) * | 2014-10-17 | 2015-01-07 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
CN104299229A (en) * | 2014-09-23 | 2015-01-21 | 西安电子科技大学 | Infrared weak and small target detection method based on time-space domain background suppression |
CN104809702A (en) * | 2015-04-22 | 2015-07-29 | 上海理工大学 | Pulse diagnosis curve grid eliminating method based on frequency domain processing |
CN104834915A (en) * | 2015-05-15 | 2015-08-12 | 中国科学院武汉物理与数学研究所 | Small infrared object detection method in complex cloud sky background |
US20150248579A1 (en) * | 2013-12-24 | 2015-09-03 | Huazhong University Of Science And Technology | Method for identifying and positioning building using outline region restraint of mountain |
CN106251344A (en) * | 2016-07-26 | 2016-12-21 | 北京理工大学 | A kind of multiple dimensioned infrared target self-adapting detecting method of view-based access control model receptive field |
CN107194355A (en) * | 2017-05-24 | 2017-09-22 | 北京航空航天大学 | A kind of utilization orientation derivative constructs the method for detecting infrared puniness target of entropy contrast |
CN107590496A (en) * | 2017-09-18 | 2018-01-16 | 南昌航空大学 | The association detection method of infrared small target under complex background |
CN108198198A (en) * | 2017-12-22 | 2018-06-22 | 湖南源信光电科技股份有限公司 | Single frames infrared small target detection method based on wavelet transformation and Steerable filter |
-
2018
- 2018-08-21 CN CN201810956335.9A patent/CN109272489B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5384716A (en) * | 1992-09-24 | 1995-01-24 | Matsushita Electric Industrial Co., Ltd. | Occupant condition determining apparatus |
CN102819740A (en) * | 2012-07-18 | 2012-12-12 | 西北工业大学 | Method for detecting and positioning dim targets of single-frame infrared image |
CN103632370A (en) * | 2013-12-02 | 2014-03-12 | 西北工业大学 | Multilevel iteration visualization optimization based total-focusing synthetic aperture perspective imaging method |
US20150248579A1 (en) * | 2013-12-24 | 2015-09-03 | Huazhong University Of Science And Technology | Method for identifying and positioning building using outline region restraint of mountain |
CN104299229A (en) * | 2014-09-23 | 2015-01-21 | 西安电子科技大学 | Infrared weak and small target detection method based on time-space domain background suppression |
CN104268844A (en) * | 2014-10-17 | 2015-01-07 | 中国科学院武汉物理与数学研究所 | Small target infrared image processing method based on weighing local image entropy |
CN104809702A (en) * | 2015-04-22 | 2015-07-29 | 上海理工大学 | Pulse diagnosis curve grid eliminating method based on frequency domain processing |
CN104834915A (en) * | 2015-05-15 | 2015-08-12 | 中国科学院武汉物理与数学研究所 | Small infrared object detection method in complex cloud sky background |
CN106251344A (en) * | 2016-07-26 | 2016-12-21 | 北京理工大学 | A kind of multiple dimensioned infrared target self-adapting detecting method of view-based access control model receptive field |
CN107194355A (en) * | 2017-05-24 | 2017-09-22 | 北京航空航天大学 | A kind of utilization orientation derivative constructs the method for detecting infrared puniness target of entropy contrast |
CN107590496A (en) * | 2017-09-18 | 2018-01-16 | 南昌航空大学 | The association detection method of infrared small target under complex background |
CN108198198A (en) * | 2017-12-22 | 2018-06-22 | 湖南源信光电科技股份有限公司 | Single frames infrared small target detection method based on wavelet transformation and Steerable filter |
Non-Patent Citations (4)
Title |
---|
CHEN YUWEN等: "《An effective infrared small target detection method based on the human visual attention》", 《INFRARED PHYSICS AND TECHNOLOGY》 * |
关如冰: "《多阶段的红外小目标检测算法》", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
周伟: "《视觉注意机制下的红外弱小目标显著性检测》", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
武斌 等: "《一种新的红外弱小目标检测前跟踪算法》", 《西安电子科技大学学报》 * |
Cited By (28)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109903272A (en) * | 2019-01-30 | 2019-06-18 | 西安天伟电子系统工程有限公司 | Object detection method, device, equipment, computer equipment and storage medium |
CN109949292A (en) * | 2019-03-20 | 2019-06-28 | 湖南镭目科技有限公司 | A kind of detection method and device of steel strip surface defect |
CN109949292B (en) * | 2019-03-20 | 2023-11-14 | 湖南镭目科技有限公司 | Method and device for detecting surface defects of strip steel |
CN110135312B (en) * | 2019-05-06 | 2022-05-03 | 电子科技大学 | Rapid small target detection method based on hierarchical LCM |
CN110135312A (en) * | 2019-05-06 | 2019-08-16 | 电子科技大学 | A kind of quick small target detecting method based on classification LCM |
CN110490848A (en) * | 2019-08-02 | 2019-11-22 | 上海海事大学 | Infrared target detection method, apparatus and computer storage medium |
CN110717499A (en) * | 2019-09-24 | 2020-01-21 | 南京理工大学 | Image feature extraction method based on single-layer perceptron model |
CN110717499B (en) * | 2019-09-24 | 2022-08-09 | 南京理工大学 | Image feature extraction method based on single-layer perceptron model |
CN110674782A (en) * | 2019-09-30 | 2020-01-10 | 电子科技大学 | Infrared dim target detection method based on fractional entropy |
CN111160260A (en) * | 2019-12-30 | 2020-05-15 | 湖北航天技术研究院总体设计所 | SAR image target detection method and system |
CN111160260B (en) * | 2019-12-30 | 2023-07-04 | 湖北航天技术研究院总体设计所 | SAR image target detection method and system |
CN111260624B (en) * | 2020-01-15 | 2023-05-26 | 陕西师范大学 | Multi-scale infrared small target detection method based on feature matching and scale selection strategy |
CN111260624A (en) * | 2020-01-15 | 2020-06-09 | 陕西师范大学 | Multi-scale infrared small target detection method based on feature matching and scale selection strategy |
CN111369458A (en) * | 2020-02-28 | 2020-07-03 | 中国人民解放军空军工程大学 | Infrared dim target background suppression method based on multi-scale rolling guide filtering smoothing |
CN111369458B (en) * | 2020-02-28 | 2023-04-07 | 中国人民解放军空军工程大学 | Infrared dim target background suppression method based on multi-scale rolling guide filtering smoothing |
CN112163606A (en) * | 2020-09-17 | 2021-01-01 | 哈尔滨工业大学(威海) | Infrared small target detection method based on block contrast weighting |
CN112766109B (en) * | 2021-01-08 | 2022-07-22 | 北京航空航天大学 | Infrared weak and small target self-adaptive detection method |
CN112766109A (en) * | 2021-01-08 | 2021-05-07 | 北京航空航天大学 | Infrared weak and small target self-adaptive detection method |
CN113076888A (en) * | 2021-04-09 | 2021-07-06 | 中国电子科技集团公司第二十八研究所 | Multi-scale local binary model photoelectric image target detection method |
CN113205494B (en) * | 2021-04-27 | 2022-05-13 | 武汉大学 | Infrared small target detection method and system based on adaptive scale image block weighting difference measurement |
CN113205494A (en) * | 2021-04-27 | 2021-08-03 | 武汉大学 | Infrared small target detection method and system based on adaptive scale image block weighting difference measurement |
CN113658191A (en) * | 2021-07-05 | 2021-11-16 | 中国人民解放军火箭军工程大学 | Infrared dim target detection method based on local probability hypergraph dissimilarity measure |
CN113643315B (en) * | 2021-07-22 | 2023-08-15 | 武汉大学 | Infrared small target detection method based on self-adaptive peak gradient descent filter |
CN113643315A (en) * | 2021-07-22 | 2021-11-12 | 武汉大学 | Infrared small target detection method based on self-adaptive peak gradient descent filter |
CN113822352A (en) * | 2021-09-15 | 2021-12-21 | 中北大学 | Infrared dim target detection method based on multi-feature fusion |
CN113822352B (en) * | 2021-09-15 | 2024-05-17 | 中北大学 | Infrared dim target detection method based on multi-feature fusion |
CN116645580A (en) * | 2023-06-05 | 2023-08-25 | 北京邮电大学 | Method and device for detecting infrared dim and small targets based on space-time characteristic difference |
CN116645580B (en) * | 2023-06-05 | 2023-11-14 | 北京邮电大学 | Weak and small target detection method and device based on space-time characteristic difference |
Also Published As
Publication number | Publication date |
---|---|
CN109272489B (en) | 2022-03-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109272489A (en) | Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background | |
CN106384344B (en) | A kind of remote sensing image surface vessel target detection and extracting method | |
Qi et al. | A robust directional saliency-based method for infrared small-target detection under various complex backgrounds | |
CN107145874A (en) | Ship Target Detection and discrimination method in complex background SAR image | |
US8103055B2 (en) | Detection of blobs in images | |
CN106485724B (en) | It is a kind of based on combination receptive field and towards feature modulate profile testing method | |
CN106251344A (en) | A kind of multiple dimensioned infrared target self-adapting detecting method of view-based access control model receptive field | |
CN104766079B (en) | A kind of remote method for detecting infrared puniness target | |
CN104834915B (en) | A kind of small infrared target detection method under complicated skies background | |
CN109816051B (en) | Hazardous chemical cargo feature point matching method and system | |
CN107194355B (en) | A kind of method for detecting infrared puniness target of utilization orientation derivative construction entropy contrast | |
Cao et al. | Infrared small target detection based on derivative dissimilarity measure | |
CN109523575A (en) | Infrared weak and small target detection method | |
Lian et al. | A novel method on moving-objects detection based on background subtraction and three frames differencing | |
CN106056115B (en) | A kind of infrared small target detection method under non-homogeneous background | |
CN106600613B (en) | Improvement LBP infrared target detection method based on embedded gpu | |
Fu et al. | Infrared sea-sky line detection utilizing self-adaptive Laplacian of Gaussian filter and visual-saliency-based probabilistic Hough transform | |
CN108647693A (en) | A kind of sea infrared target detection method based on binary significant characteristics | |
Zhang et al. | A Detection Method with Anti-interference for Infrared Maritime Small Target | |
He et al. | An improved method MSS-YOLOv5 for object detection with balancing speed-accuracy | |
CN110310263A (en) | A kind of SAR image residential block detection method based on significance analysis and background priori | |
Zhang et al. | Infrared dim target detection method based on the fuzzy accurate updating symmetric adaptive resonance theory | |
CN116385866B (en) | SAR image-based railway line color steel house change detection method and device | |
CN105551029B (en) | A kind of multi-spectral remote sensing image Ship Detection | |
CN111089586A (en) | All-day star sensor star point extraction method based on multi-frame accumulation algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230728 Address after: Room A601, National Digital Publishing Base, No. 996 Tiangu 7th Road, Gaoxin, Xi'an City, Shaanxi Province, 710000 Patentee after: XI'AN THUNDER DRIVEN ELECTRONIC TECHNOLOGY CO.,LTD. Address before: 710071 No. 2 Taibai South Road, Shaanxi, Xi'an Patentee before: XIDIAN University |