CN109102003B - Small target detection method and system based on infrared physical characteristic fusion - Google Patents
Small target detection method and system based on infrared physical characteristic fusion Download PDFInfo
- Publication number
- CN109102003B CN109102003B CN201810790004.2A CN201810790004A CN109102003B CN 109102003 B CN109102003 B CN 109102003B CN 201810790004 A CN201810790004 A CN 201810790004A CN 109102003 B CN109102003 B CN 109102003B
- Authority
- CN
- China
- Prior art keywords
- target
- image
- marked
- infrared
- target area
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Transforming Light Signals Into Electric Signals (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a small target detection method and system based on infrared physical characteristic fusion, wherein the method comprises the following steps: marking small targets in the sample multiband infrared image to obtain marked targets, randomly selecting non-targets in the sample multiband infrared image to mark to obtain marked non-targets, and training a classifier by using the marked targets and the characteristic vectors of the marked non-targets to obtain a target classifier. Segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area; extracting the characteristic vector of the candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target. The invention improves the small target detection rate and reduces the false alarm rate of small target detection under the complex background.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a small target detection method and system based on infrared physical characteristic fusion.
Background
Detection of small objects in a complex background is a difficult problem. The small target carries a small amount of information, has low contrast and is easy to be annihilated in a complex background. The method is an important approach for improving the target detection performance by utilizing the information of infrared images with different wave bands. The method has important significance for improving the small target detection performance by fusing the information of the infrared images of different wave bands.
The target detection method based on image fusion is mainly divided into three categories, namely a small target detection method based on pixel level fusion, a small target detection method based on feature level fusion and a small target detection method based on decision level fusion. The small target detection method based on pixel level fusion adopts methods such as linear fusion, multi-resolution analysis fusion and the like to fuse medium wave and long wave infrared images, and then performs target detection on the fused images, wherein the small target detection effect of the method depends on the small target enhancement degree in the fusion strategy. Feature level fusion based methods are less useful for small object detection because small object information volumes are less likely to extract suitable features. The small target detection method based on decision-level fusion carries out small target detection in different wave bands respectively, and then fusion is carried out on the detection results to give a final detection result. The target detection performance of such methods depends on the small target detection results of different bands. The existing small target under the complex background is difficult to detect and easy to detect by mistake and miss.
Disclosure of Invention
Aiming at the defects or the improvement requirements of the prior art, the invention provides a small target detection method and a small target detection system based on infrared physical characteristic fusion, so that the technical problems of difficult small target detection, easy false detection and missed detection under the existing complex background are solved.
In order to achieve the above object, according to an aspect of the present invention, there is provided a small target detection method based on infrared physical feature fusion, including:
(1) segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
(2) extracting a characteristic vector of a candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target;
the training of the target classifier comprises: marking small targets in the sample multiband infrared image to obtain marked targets, randomly selecting non-targets in the sample multiband infrared image to mark to obtain marked non-targets, and training a classifier by using the marked targets and the characteristic vectors of the marked non-targets to obtain a target classifier.
Further, a small target is a target with a pixel scale less than 6 × 6 and a local signal-to-noise ratio less than 3.
Further, the step (1) comprises:
(11) performing morphological filtering processing on the infrared image to obtain a filtered image, and calculating a segmentation threshold of the filtered image;
(12) and utilizing a segmentation threshold value to segment the filtered image to obtain a segmented image, and marking the segmented image to obtain a candidate target area.
Further, the step (2) comprises:
(21) performing imaging inverse transformation processing on the infrared image to obtain an infrared radiation energy image, acquiring an infrared radiation energy value of a candidate target area from the infrared radiation energy image, and calculating a local signal-to-noise ratio value of a central pixel position of the candidate target area;
(22) combining the infrared radiation energy value of the candidate target area and the local signal-to-noise ratio value of the central pixel position of the candidate target area to obtain a feature vector of the candidate target area, inputting the feature vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target.
Further, the obtaining of the feature vectors of the labeled target and the labeled non-target includes:
the method comprises the steps of carrying out imaging inverse transformation processing on a sample multiband infrared image to obtain a sample multiband infrared intermediate wave radiation energy image, obtaining multiband infrared radiation energy values of a marked target and a marked non-target from the sample multiband infrared intermediate wave radiation energy image, respectively calculating local signal-to-noise values of central pixel positions of the marked target and the marked non-target, and combining the multiband infrared radiation energy values of the marked target and the marked non-target and the local signal-to-noise values of the central pixel positions of the marked target and the marked non-target to obtain characteristic vectors of the marked target and the marked non-target.
According to another aspect of the present invention, there is provided a small target detection system based on infrared physical feature fusion, including:
the classifier training module is used for marking small targets in the sample multiband infrared image to obtain marked targets, randomly selecting non-targets in the sample multiband infrared image to mark to obtain marked non-targets, and training a classifier by using the marked targets and the characteristic vectors of the marked non-targets to obtain a target classifier;
the image segmentation module is used for segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
and the target detection module is used for extracting the characteristic vector of the candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into the target classifier and detecting whether the candidate target area has a small target.
Further, a small target is a target with a pixel scale less than 6 × 6 and a local signal-to-noise ratio less than 3.
Further, the image segmentation module includes:
the image filtering submodule is used for performing morphological filtering processing on the infrared image to obtain a filtered image and calculating a segmentation threshold of the filtered image;
and the image segmentation submodule is used for segmenting the filtered image by utilizing a segmentation threshold value to obtain a segmented image, and marking the segmented image to obtain a candidate target area.
Further, the object detection module includes:
the image transformation submodule is used for carrying out imaging inverse transformation processing on the infrared image to obtain an infrared radiation energy image;
the radiation energy extraction submodule is used for acquiring the infrared radiation energy value of the candidate target area from the infrared radiation energy image;
the local signal-to-noise ratio extraction submodule is used for calculating the local signal-to-noise ratio of the central pixel position of the candidate target area;
the feature vector combination submodule is used for combining the infrared radiation energy value of the candidate target area and the local signal-to-noise ratio value of the central pixel position of the candidate target area to obtain a feature vector of the candidate target area;
and the target detection submodule is used for inputting the feature vectors of the candidate target areas into the target classifier and detecting whether the candidate target areas have small targets or not.
Further, the obtaining of the feature vectors of the labeled target and the labeled non-target includes:
the method comprises the steps of carrying out imaging inverse transformation processing on a sample multiband infrared image to obtain a sample multiband infrared intermediate wave radiation energy image, obtaining multiband infrared radiation energy values of a marked target and a marked non-target from the sample multiband infrared intermediate wave radiation energy image, respectively calculating local signal-to-noise values of central pixel positions of the marked target and the marked non-target, and combining the multiband infrared radiation energy values of the marked target and the marked non-target and the local signal-to-noise values of the central pixel positions of the marked target and the marked non-target to obtain characteristic vectors of the marked target and the marked non-target.
In general, compared with the prior art, the above technical solution contemplated by the present invention can achieve the following beneficial effects:
(1) the method comprises the steps of firstly segmenting an infrared image, marking the segmented image and preliminarily detecting a candidate target area; then extracting the characteristic vector of the candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target; the small target is detected by adopting a coarse and fine two-stage processing mode, and meanwhile, the characteristic vector of a candidate target area in the infrared image is used for distinguishing, so that the false alarm rate of small target detection can be reduced, and the small target detection rate is improved.
(2) The feature vector of the invention fuses the infrared radiation energy value and the local signal-to-noise ratio value, and can improve the small target information quantity and further improve the small target detection rate. During training, carrying out inverse transformation on the multiband infrared image to obtain a multiband infrared radiation energy image, and extracting an infrared radiation energy value and a local signal-to-noise ratio value of the multiband infrared image to form a feature vector; compared with the prior art, the method utilizes the multiband infrared physical radiation characteristics through the characteristic level fusion method, so that the small target information quantity can be increased, and the small target detection rate can be increased.
Drawings
Fig. 1 is a flowchart of a small target detection method based on infrared physical feature fusion according to an embodiment of the present invention;
FIG. 2(a) is a medium wave infrared image provided in example 1 of the present invention;
FIG. 2(b) is a long-wave infrared image provided in example 1 of the present invention;
FIG. 3 is an image of a marked target and a non-target provided in embodiment 1 of the present invention;
FIG. 4 shows the results of the preliminary test provided in example 1 of the present invention;
fig. 5 is a result of detecting a small target provided in embodiment 1 of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
As shown in fig. 1, a small target detection method based on infrared physical feature fusion includes:
(1) segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
(2) extracting a characteristic vector of a candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target;
the training of the target classifier comprises: marking small targets in the sample multiband infrared image to obtain marked targets, randomly selecting non-targets in the sample multiband infrared image to mark to obtain marked non-targets, and training a classifier by using the marked targets and the characteristic vectors of the marked non-targets to obtain a target classifier.
Example 1
A small target detection method based on infrared physical characteristic fusion comprises the following steps:
(1) segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
(2) extracting a characteristic vector of a candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target;
the training of the target classifier comprises: the medium wave infrared image shown in fig. 2(a) and the long wave infrared image shown in fig. 2(b) are subjected to imaging inverse transformation processing according to the formulaWherein g is a medium wave infrared image or a long wave infrared image, and K and B are conversion parameters calibrated by using a black body to obtain an infrared medium wave and long wave radiation energy image pair. As shown in fig. 3, small targets in the infrared medium wave and long wave radiation energy image pairs are marked to obtain marked targets, and non-targets in the infrared medium wave and long wave radiation energy image pairs are randomly selected to be marked to obtain marked non-targets.
Extracting L mid-wave IR energy values for pixels (x, y) in tagged targets and tagged non-targetsM(x, y) and long wave infrared radiation energy value LL(x,y)。
Respectively calculating the local signal-to-noise ratio of the pixel (x, y) according to the formulaThe signal-to-noise ratio of the local area 21 × 21 centered on pixel (x, y) was calculated to obtain the SNRM(x, y) and SNRL(x,y);μTRepresenting the mean value of the target pixel gray level, muBRepresenting the mean value of the background pixel intensity, σ, in the neighborhood of 21 × 21 centered on the targetBIs the mean square error of the pixel gray levels in the neighborhood range.
Combining the extracted mid-wave and long-wave infrared radiation energy values of the pixel (x, y) and the value of the local signal-to-noise ratio into a feature vector EV (x, y) ((L))M(x,y),LL(x,y),SNRM(x,y),SNRL(x,y))。
Specifically, the step (1) comprises the following steps:
(11) performing morphological filtering processing on the infrared image to obtain a filtered image, and calculating a segmentation threshold of the filtered image; the calculation formula of the segmentation threshold is Th (mu + k sigma), mu is a mean value after filtering, sigma is a difference after filtering, and the parameter k is adjustable;
(12) the filtered image is segmented by using a segmentation threshold to obtain a segmented image, and the segmented image is marked to obtain a plurality of candidate target regions, which are shown as regions marked as white frames in fig. 4.
Specifically, the step (2) includes:
(21) performing imaging inverse transformation processing on the infrared image to obtain an infrared radiation energy image, acquiring an infrared radiation energy value of each candidate target area from the infrared radiation energy image, and calculating a local signal-to-noise ratio of a central pixel position of each candidate target area;
(22) combining the infrared radiation energy value of each candidate target area with the local signal-to-noise ratio value of the central pixel position of the corresponding candidate target area to obtain a feature vector of each candidate target area, inputting the feature vector of each candidate target area into a target classifier, and detecting whether each candidate target area has a small target or not to obtain a final small target detection result, which is shown as a white frame marked area in fig. 5.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (6)
1. A small target detection method based on infrared physical characteristic fusion is characterized by comprising the following steps:
(1) segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
(2) extracting a characteristic vector of a candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target;
the training of the target classifier comprises: marking small targets in the sample multiband infrared image to obtain marked targets, randomly selecting non-targets in the sample multiband infrared image to mark to obtain marked non-targets, and training a classifier by using the marked targets and the characteristic vectors of the marked non-targets to obtain a target classifier;
the step (1) comprises the following steps:
(11) performing morphological filtering processing on the infrared image to obtain a filtered image, and calculating a segmentation threshold of the filtered image;
(12) segmenting the filtered image by utilizing a segmentation threshold value to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
the step (2) comprises the following steps:
(21) performing imaging inverse transformation processing on the infrared image to obtain an infrared radiation energy image, acquiring an infrared radiation energy value of a candidate target area from the infrared radiation energy image, and calculating a local signal-to-noise ratio value of a central pixel position of the candidate target area;
(22) combining the infrared radiation energy value of the candidate target area and the local signal-to-noise ratio value of the central pixel position of the candidate target area to obtain a feature vector of the candidate target area, inputting the feature vector of the candidate target area into a target classifier, and detecting whether the candidate target area has a small target.
2. The method for detecting the small target based on the infrared physical feature fusion as claimed in claim 1, wherein the small target is a target with a pixel scale less than 6 × 6 and a local signal-to-noise ratio less than 3.
3. The method for detecting the small target based on the infrared physical feature fusion as claimed in claim 1 or 2, wherein the obtaining of the feature vectors of the marked target and the marked non-target comprises:
the method comprises the steps of carrying out imaging inverse transformation processing on a sample multiband infrared image to obtain a sample multiband infrared radiation energy image, obtaining multiband infrared radiation energy values of a marked target and a marked non-target from the sample multiband infrared radiation energy image, respectively calculating local signal-to-noise values of central pixel positions of the marked target and the marked non-target, and combining the multiband infrared radiation energy values of the marked target and the marked non-target and the local signal-to-noise values of the central pixel positions of the marked target and the marked non-target to obtain a characteristic vector of the marked target and the marked non-target.
4. A small target detection system based on infrared physical feature fusion is characterized by comprising:
the classifier training module is used for marking small targets in the sample multiband infrared image to obtain marked targets, randomly selecting non-targets in the sample multiband infrared image to mark to obtain marked non-targets, and training a classifier by using the marked targets and the characteristic vectors of the marked non-targets to obtain a target classifier;
the image segmentation module is used for segmenting the infrared image to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
the target detection module is used for extracting the characteristic vector of the candidate target area in the infrared image, inputting the characteristic vector of the candidate target area into the target classifier and detecting whether the candidate target area has a small target;
the image segmentation module comprises:
the image filtering submodule is used for performing morphological filtering processing on the infrared image to obtain a filtered image and calculating a segmentation threshold of the filtered image;
the image segmentation submodule is used for segmenting the filtered image by utilizing a segmentation threshold value to obtain a segmented image, and marking the segmented image to obtain a candidate target area;
the target detection module includes:
the image transformation submodule is used for carrying out imaging inverse transformation processing on the infrared image to obtain an infrared radiation energy image;
the radiation energy extraction submodule is used for acquiring the infrared radiation energy value of the candidate target area from the infrared radiation energy image;
the local signal-to-noise ratio extraction submodule is used for calculating the local signal-to-noise ratio of the central pixel position of the candidate target area;
the feature vector combination submodule is used for combining the infrared radiation energy value of the candidate target area and the local signal-to-noise ratio value of the central pixel position of the candidate target area to obtain a feature vector of the candidate target area;
and the target detection submodule is used for inputting the feature vectors of the candidate target areas into the target classifier and detecting whether the candidate target areas have small targets or not.
5. The infrared physical feature fusion-based small target detection system as claimed in claim 4, wherein the small target is a target with a pixel scale of less than 6 × 6 and a local signal-to-noise ratio of less than 3.
6. The infrared physical feature fusion-based small target detection system as claimed in claim 4 or 5, wherein the obtaining of the feature vectors of the labeled target and the labeled non-target comprises:
the method comprises the steps of carrying out imaging inverse transformation processing on a sample multiband infrared image to obtain a sample multiband infrared radiation energy image, obtaining multiband infrared radiation energy values of a marked target and a marked non-target from the sample multiband infrared radiation energy image, respectively calculating local signal-to-noise values of central pixel positions of the marked target and the marked non-target, and combining the multiband infrared radiation energy values of the marked target and the marked non-target and the local signal-to-noise values of the central pixel positions of the marked target and the marked non-target to obtain a characteristic vector of the marked target and the marked non-target.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810790004.2A CN109102003B (en) | 2018-07-18 | 2018-07-18 | Small target detection method and system based on infrared physical characteristic fusion |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810790004.2A CN109102003B (en) | 2018-07-18 | 2018-07-18 | Small target detection method and system based on infrared physical characteristic fusion |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109102003A CN109102003A (en) | 2018-12-28 |
CN109102003B true CN109102003B (en) | 2020-07-10 |
Family
ID=64846711
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810790004.2A Active CN109102003B (en) | 2018-07-18 | 2018-07-18 | Small target detection method and system based on infrared physical characteristic fusion |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109102003B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111401195B (en) * | 2020-03-10 | 2023-11-14 | 上海航天控制技术研究所 | Sea surface target detection method based on multiband infrared image |
CN111291762B (en) * | 2020-03-10 | 2022-12-13 | 上海航天控制技术研究所 | Multi-feature-point-difference-based multi-band image fusion detection method |
CN112837335B (en) * | 2021-01-27 | 2023-05-09 | 上海航天控制技术研究所 | Medium-long wave infrared composite anti-interference method |
CN113450413B (en) * | 2021-07-19 | 2022-09-27 | 哈尔滨工业大学 | Ship target detection method based on GF4 single-frame image |
CN114463619B (en) * | 2022-04-12 | 2022-07-08 | 西北工业大学 | Infrared dim target detection method based on integrated fusion features |
CN115830470B (en) * | 2022-12-29 | 2024-07-09 | 中国科学院长春光学精密机械与物理研究所 | Method, device and equipment for detecting weak-intensity small-scale target of remote sensing image |
CN116486086B (en) * | 2023-04-28 | 2023-10-03 | 安徽星太宇科技有限公司 | Target detection method based on thermal infrared remote sensing image |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102034239A (en) * | 2010-12-07 | 2011-04-27 | 北京理工大学 | Local gray abrupt change-based infrared small target detection method |
US7965890B2 (en) * | 2007-01-05 | 2011-06-21 | Raytheon Company | Target recognition system and method |
CN104899866A (en) * | 2015-05-05 | 2015-09-09 | 河南三联网络技术有限公司 | Intelligent infrared small target detection method |
-
2018
- 2018-07-18 CN CN201810790004.2A patent/CN109102003B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7965890B2 (en) * | 2007-01-05 | 2011-06-21 | Raytheon Company | Target recognition system and method |
CN102034239A (en) * | 2010-12-07 | 2011-04-27 | 北京理工大学 | Local gray abrupt change-based infrared small target detection method |
CN104899866A (en) * | 2015-05-05 | 2015-09-09 | 河南三联网络技术有限公司 | Intelligent infrared small target detection method |
Non-Patent Citations (2)
Title |
---|
基于多级综合分类器的红外目标检测;王培早 等;《软件工程》;20171231;第20卷(第12期);第2-5页 * |
基于特征级融合的多波段舰船目标识别方法;刘峰 等;《光谱学与光谱分析》;20170630;第37卷(第6期);第1934页 * |
Also Published As
Publication number | Publication date |
---|---|
CN109102003A (en) | 2018-12-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109102003B (en) | Small target detection method and system based on infrared physical characteristic fusion | |
Zhang et al. | GAN-FM: Infrared and visible image fusion using GAN with full-scale skip connection and dual Markovian discriminators | |
CN111292305B (en) | Improved YOLO-V3 metal processing surface defect detection method | |
CN107808378B (en) | Method for detecting potential defects of complex-structure casting based on vertical longitudinal and transverse line profile features | |
CN108985345B (en) | Detection apparatus based on lung medical image fusion classification | |
CN108573222B (en) | Pedestrian image occlusion detection method based on cyclic confrontation generation network | |
CN102509098B (en) | Fisheye image vehicle identification method | |
CN111028206A (en) | Prostate cancer automatic detection and classification system based on deep learning | |
CN111611907B (en) | Image-enhanced infrared target detection method | |
CN106530271B (en) | A kind of infrared image conspicuousness detection method | |
TW201120806A (en) | Method of locating license plate of moving vehicle | |
US11538261B2 (en) | Systems and methods for automated cell segmentation and labeling in immunofluorescence microscopy | |
CN110659649A (en) | Image processing and character recognition algorithm based on near infrared light imaging | |
CN109975196B (en) | Reticulocyte detection method and system | |
CN109559324A (en) | A kind of objective contour detection method in linear array images | |
CN106127728A (en) | A kind of infrared image sea horizon connected domain detection method under sea and sky background | |
CN109978848A (en) | Method based on hard exudate in multiple light courcess color constancy model inspection eye fundus image | |
CN109544535B (en) | Peeping camera detection method and system based on optical filtering characteristics of infrared cut-off filter | |
CN111738211B (en) | PTZ camera moving object detection and recognition method based on dynamic background compensation and deep learning | |
CN110675442B (en) | Local stereo matching method and system combined with target recognition technology | |
CN104036250A (en) | Video pedestrian detecting and tracking method | |
CN109447119A (en) | Cast recognition methods in the arena with SVM is cut in a kind of combining form credit | |
CN109165636A (en) | A kind of sparse recognition methods of Rare Birds based on component-level multiple features fusion | |
CN110458019B (en) | Water surface target detection method for eliminating reflection interference under scarce cognitive sample condition | |
CN111476074A (en) | Human body foreign matter detection method based on millimeter wave image |
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 |