CN111695398A - Small target ship identification method and device and electronic equipment - Google Patents
Small target ship identification method and device and electronic equipment Download PDFInfo
- Publication number
- CN111695398A CN111695398A CN201911350054.XA CN201911350054A CN111695398A CN 111695398 A CN111695398 A CN 111695398A CN 201911350054 A CN201911350054 A CN 201911350054A CN 111695398 A CN111695398 A CN 111695398A
- Authority
- CN
- China
- Prior art keywords
- ship
- image
- small target
- small
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of computer vision of intelligent technology, in particular to a method and a device for identifying a small target ship and electronic equipment. The method comprises the following steps: acquiring ship data; carrying out image amplification preprocessing according to ship data to obtain a preprocessed ship image; inputting a target detection and identification network based on a training ship image to construct a small target ship identification model; and inputting the verification ship image into the constructed small target ship identification model for detection and outputting the small target ship identification result. The small target data image is extracted by preprocessing the acquired ship data set, the small target data image is subjected to image amplification, the screened small target image is identified by utilizing a learning network, the ship image data is input into the trained identification model by utilizing an iterative learning network to obtain the identified small target ship image. Therefore, the problem that the ship features are not obvious and cannot be identified due to the fact that the ship is displayed too small in the camera or the fishing ship is too small when the ship is driven from a far place is solved.
Description
Technical Field
The invention relates to the field of computer vision of intelligent technology, in particular to a method and a device for identifying a small target ship and electronic equipment.
Background
In the current electronic purse net business, the ship features are not obvious and can not be identified because the ship is displayed too small in a camera or a fishing boat is too small when the ship drives from a distance, and further the electronic purse net brings difficulty to the monitoring of the ship at sea.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method, an apparatus, and an electronic device for identifying a small target ship, so as to solve the problem that when a ship arrives from a distant place, the ship features are not obvious and cannot be identified due to too small display in a camera or too small presence of a fishing boat.
According to a first aspect, an embodiment of the present invention provides a small target ship identification method, including:
acquiring ship data;
carrying out image amplification preprocessing according to the ship data to obtain a preprocessed ship image; wherein the pre-processing the ship image comprises: training ship images and verifying ship images;
inputting a target detection and identification network based on the training ship image, and constructing a small target ship identification model;
and inputting the verification ship image into the constructed small target ship identification model for detection, and outputting a small target ship identification result.
The method comprises the steps of preprocessing an acquired ship data set, extracting a small target data image, amplifying the small target data image, identifying the screened small target image data by using a learning network, obtaining an identification model by using an iterative learning network, and inputting ship image data into a trained identification model to obtain the identified small target ship image. Therefore, the problem that the ship features are not obvious and cannot be identified due to the fact that the ship is displayed too small in the camera or the fishing ship is too small when the ship is driven from a far place is solved.
With reference to the first aspect, in a first embodiment of the first aspect, the image enlargement preprocessing includes:
screening small-size images based on the acquired ship data;
and utilizing the Laplacian pyramid to perform upsampling on the screened small-size image so as to obtain a large-size image.
The method comprises the steps of screening acquired ship data, distinguishing identifiable large target data from unidentified small target ship data, carrying out a method on a small target ship image, enabling the characteristics of the small target ship to be detected, enabling the data to be amplified by utilizing up-sampling, ensuring the characteristic information of original ship data as far as possible while avoiding losing key characteristics due to over-sampling, and therefore preprocessing a ship data set is carried out to ensure that the small target image can be identified in order to ensure that the ship image is subsequently sent into a network, so that the small target ship can be identified, and the method is favorable for monitoring ships by electronic purse net services.
With reference to the first aspect, in a second embodiment of the first aspect, the laplacian pyramid image is:
wherein G isiExpressed as the ith layer of the image whose original size is small, UP is upsampling, g5×5A gaussian convolution kernel of 5X 5.
By using the Laplace pyramid image to sample the small-size image, the clarity of the image is guaranteed, and the small target image can be accurately identified by a subsequent network.
With reference to the first aspect, in a third implementation manner of the first aspect, the object detection and identification network includes:
inputting the training ship image into a convolutional layer for feature extraction, and outputting a feature map of the ship image;
inputting the feature map of the ship image into an RPN to extract a ship image candidate area;
extracting a feature map of the ship image candidate area by utilizing an ROI pooling layer;
and inputting the characteristic diagram of the ship image candidate region into a Softmax layer for classification, using the characteristic diagram for a frame regression full-connection layer for correction, and outputting a small target ship identification result.
With reference to the first aspect, in a fourth embodiment of the first aspect, constructing a small target vessel recognition model includes: and circularly iterating the target identification detection network until the frame regresses to obtain the small target ship identification result.
Through the small target ship identification model, the small target ship data can be accurately identified. And finally, a Softmax layer and data iteration are utilized to enable the output small target recognition result to be more accurate.
According to a second aspect, an embodiment of the present invention provides a small target ship recognition apparatus, including:
the acquisition module is used for acquiring ship data;
the preprocessing module is used for carrying out image amplification preprocessing according to the ship data to obtain a preprocessed ship image; wherein the pre-processing the ship image comprises: training ship images and verifying ship images;
the training module is used for inputting a target detection and identification network based on the training ship image and constructing a small target ship identification model;
and the identification module is used for inputting the verification ship image into the constructed small target ship identification model for detection and outputting a small target ship identification result.
By utilizing the data transmission between the modules, the identification of the small target ship can be ensured.
According to a third aspect, an embodiment of the present invention provides an electronic device, including: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute the small target vessel identification method according to the first aspect or any one of the embodiments of the first aspect.
According to a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the small target vessel identification method according to the first aspect or any one of the embodiments of the first aspect.
Drawings
The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are illustrative and not to be construed as limiting the invention in any way, and in which:
FIG. 1 is a flow chart of a method of small target vessel identification according to an embodiment of the present invention;
FIG. 2 is a flow chart of a preprocessing of a method of identifying small target vessels according to an embodiment of the invention;
FIG. 3 is a flow chart of a target identification network for a method of identifying small target vessels, according to an embodiment of the present invention;
fig. 4 is a schematic structural view of a small target ship recognition apparatus according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of an electronic device according to an embodiment of the invention;
reference numerals
10-an acquisition module; 20-a pre-processing module; 30-a training module; 40-an identification module;
21-a memory; 22-a processor; 23-bus.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a method for identifying a small target ship, which comprises the following steps of:
s10, acquiring ship data; ship picture data, particularly ship data, can be obtained from various existing open source data sets, and ship monitoring video data can also be obtained from the electronic purse net.
S11, carrying out image amplification preprocessing according to the ship data to obtain a preprocessed ship image; the preprocessing may also be scaling of the image data, and enlargement and reduction of the image data.
S12, inputting a target detection and identification network based on the training ship image, and constructing a small target ship identification model; before the small target ship identification model is built, the small target ship identification needs to be iteratively tested, and the accuracy of the small target ship identification result is guaranteed.
And S13, inputting the verification ship image into the small target ship identification model for detection, and outputting the small target ship identification result.
Wherein preprocessing the image of the ship comprises: training ship images and verifying ship images; the ship image is divided into a training ship image and a verification ship image, so that the accuracy of constructing a small target ship recognition model can be ensured, the training ship image is data used in the training recognition model, and the verification ship image is used in the recognition of the small target ship and the detection of a target detection recognition network; optionally, the ratio of the training ship image to the verification ship image is 7:3, and the more training ship images and the later verification ship image, the more accurate the result of the small target ship identification is.
The acquired ship data set is preprocessed, so that the pixels of the acquired image become small, and the small-pixel image is trained and detected in real time by utilizing deep learning to obtain a small-target detection result. Therefore, the problem that ship features are not obvious and cannot be identified due to the fact that the ship is displayed too small in the camera or the fishing ship is too small when the ship is driven from a far place is solved.
The embodiment of the invention provides a method for identifying a small target ship, which is specifically shown in fig. 2 and comprises the following steps:
s21, preprocessing the acquired ship data:
screening small-size images based on the acquired ship data; the small-size images needing to be detected are selected from the screened acquired image data, or primary detection is carried out on the acquired ship data by utilizing a target identification algorithm and a classifier to classify the ship data which cannot be identified, and upsampling is utilized to amplify the image data, so that the probability of identifying the small target ships is increased. The selected small-size image may be up-sampled by using the laplacian pyramid to obtain a large-size image.
Wherein, the Laplace pyramid image is:
wherein G isiExpressed as the ith layer of the image whose original size is small, UP is upsampling, g5×5A gaussian convolution kernel of 5X 5.
And reconstructing an upper-layer non-sampled image by using a lower-layer image through the Laplace pyramid, and restoring and optimizing image data to the maximum extent on the image so as to ensure that accurate small-target ship identification can be performed.
S22, inputting the preprocessed image into the target detection and recognition network, as shown in fig. 3, the specific steps are:
s221, inputting the training ship image into the convolutional layer for feature extraction, and outputting a feature map of the ship image; for example: in VGG-16, there are 13 conv layers, 13 relu layers and 4 pooling layers in the convolutional layer part. All convolutional layers use a 3 × 3 convolutional kernel with a step size of 1, and padding is performed on the edges to 1. In this way, the convolution of the input W × W image results in an output size of (W-3+2 padding)/1+ 1W, i.e., the size of the convolution layer image does not decrease. And the pooling layer uses 2 x 2 pooling units with a step size of 2. For a W × W image, the output size is (W-2)/2+1 — W/2 after passing through the pooling layer, and the size of the image passing through one pooling layer becomes 1/2 before input.
While 13 conv layers of the convolutional layer do not change the size of the image, with 4 pooling layers, each pooling layer reducing the input to 1/2, for a W × W input, the width and height of the feature map output by the convolutional layer is W/16 × W/16, i.e., 1/16 of the input size in proportion to the width and height of the original input image. Then, each point in the feature map is calculated to correspond to an area of the original image.
S222, inputting the feature map of the ship image into an RPN to extract a ship image candidate area; s223, extracting a feature map of the ship image candidate area by using the ROI pooling layer;
and S224, inputting the feature map of the ship image candidate region into a Softmax layer for classification and used for a border regression full-connected layer for correction.
Specifically, the RPN inputs a feature map of an image extracted from a convolutional layer, where the output is divided into two parts: the position information of the candidate region and the category corresponding to the candidate region. In order to obtain the two outputs, the position information of the candidate region in the original input image and the feature map corresponding to each candidate region need to be obtained from the input feature map for classification. By downsampling of the pooling layer, points in the feature map are mapped back to the original image, and the corresponding points are not certain pixel points but rectangular areas. Each point in the feature map is mapped back to the original image through a simple mode, so that a candidate region is obtained, or the points in each feature map can be mapped onto the original image through the Faster R-CNN network, and rectangular regions with different shapes and areas are taken as the candidate regions in the original image by taking the mapped positions as the centers.
And S23, circularly iterating the target identification detection network until the frame regresses to obtain the small target ship identification result.
And the detection result is more accurate through the iterative training result. The target detection and identification network is utilized to ensure that the data ship can be quickly identified, and the predicted result is more accurate by setting a loss function.
Alternatively, the candidate regions may be 3 different dimensions 128 × 128,256 × 256,512 × 512, or may have different aspect ratios W: H ═ 1:1,1:2,2: 1.
An embodiment of the present invention provides a ship number statistics apparatus, as shown in fig. 4, including:
the acquisition module 10 is used for acquiring ship data;
the preprocessing module 20 is used for carrying out image amplification preprocessing according to ship data to obtain a preprocessed ship image; wherein the pre-processing the ship image comprises: training ship images and verifying ship images;
the training module 30 is used for inputting a target detection and identification network based on a training ship image and constructing a small target ship identification model;
and the identification module 40 is used for inputting the verification ship image into the constructed small target ship identification model for detection and outputting the small target ship identification result.
The ship data is acquired by the acquisition module 10, and the acquired data is preprocessed by the preprocessing module 20, amplified, scaled and input into the improved network module for training, and the maximum number of iterations are performed to obtain an accurate small target ship identification result. Therefore, the camera in the electronic purse net service can identify the fishing boat with too small display when the boat drives from a far place.
An embodiment of the present invention further provides an electronic device, as shown in fig. 5, the electronic device may include a processor 22 and a memory 21, where the processor 22 and the memory 21 may be connected through a bus 23 or in another manner, and fig. 5 takes the connection through the bus as an example.
The processor 22 may be a Central Processing Unit (CPU). The Processor 22 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or combinations thereof.
The memory 21, which is a non-transitory computer-readable storage medium, may be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions/modules (e.g., the obtaining module 10, the preprocessing module 20, the training module 30, and the recognition module 40 shown in fig. 4) corresponding to the key shielding method of the vehicle-mounted display device in the embodiment of the present invention. The processor 22 executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory 21, namely, implements the small target ship identification method in the above-described method embodiment.
The memory 21 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 22, and the like. Further, the memory 21 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 21 may optionally include memory located remotely from processor 22, and these remote memories may be connected to processor 22 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
One or more modules are stored in the memory 21 and when executed by the processor 22 perform the small target vessel identification method as in the embodiment shown in fig. 1-2.
The details of the above-mentioned apparatus can be understood by referring to the corresponding descriptions and effects of the embodiments shown in fig. 1 to fig. 2, which are not described herein again.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, and the program can be stored in a computer readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), a Solid State Drive (SSD), or the like; the storage medium may also comprise a combination of memories of the kind described above.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.
Claims (8)
1. A method for identifying small target vessels, comprising:
acquiring ship data;
carrying out image amplification preprocessing according to the ship data to obtain a preprocessed ship image; wherein the pre-processing the ship image comprises: training ship images and verifying ship images;
inputting a target detection and identification network based on the training ship image, and constructing a small target ship identification model;
and inputting the verification ship image into the constructed small target ship identification model for detection, and outputting a small target ship identification result.
2. The method of claim 1, wherein the image magnification pre-processing comprises:
screening small-size images based on the acquired ship data;
and utilizing the Laplacian pyramid to perform upsampling on the screened small-size image so as to obtain a large-size image.
3. The method of claim 2, wherein the laplacian pyramid image is:
wherein G isiExpressed as the ith layer of the image whose original size is small, UP is upsampling, g5×5A gaussian convolution kernel of 5X 5.
4. The method of claim 3, wherein the object detection recognition network comprises:
inputting the training ship image into a convolutional layer for feature extraction, and outputting a feature map of the ship image;
inputting the feature map of the ship image into an RPN to extract a ship image candidate area;
extracting a feature map of the ship image candidate area by utilizing an ROI pooling layer;
and inputting the characteristic diagram of the ship image candidate region into a Softmax layer for classification, using the characteristic diagram for a frame regression full-connection layer for correction, and outputting a small target ship identification result.
5. The method of claim 4, wherein the constructing a small target vessel identification model comprises: and circularly iterating the target identification detection network until the frame regresses to obtain the small target ship identification result.
6. A small target vessel identification device, further characterized by comprising:
the acquisition module is used for acquiring ship data;
the preprocessing module is used for carrying out image amplification preprocessing according to the ship data to obtain a preprocessed ship image; wherein the pre-processing the ship image comprises: training ship images and verifying ship images;
the training module is used for inputting a target detection and identification network based on the training ship image and constructing a small target ship identification model;
and the identification module is used for inputting the verification ship image into the constructed small target ship identification model for detection and outputting a small target ship identification result.
7. An electronic device, comprising:
a memory and a processor, the memory and the processor being communicatively connected to each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the small target vessel identification method of any one of claims 1 to 5.
8. A computer-readable storage medium storing computer instructions for causing a computer to perform the small target vessel identification method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911350054.XA CN111695398A (en) | 2019-12-24 | 2019-12-24 | Small target ship identification method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911350054.XA CN111695398A (en) | 2019-12-24 | 2019-12-24 | Small target ship identification method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111695398A true CN111695398A (en) | 2020-09-22 |
Family
ID=72476238
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911350054.XA Pending CN111695398A (en) | 2019-12-24 | 2019-12-24 | Small target ship identification method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111695398A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101282A (en) * | 2020-09-25 | 2020-12-18 | 北京瞰天科技有限公司 | Aquatic target identification method and device, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108122003A (en) * | 2017-12-19 | 2018-06-05 | 西北工业大学 | A kind of Weak target recognition methods based on deep neural network |
| CN109472298A (en) * | 2018-10-19 | 2019-03-15 | 天津大学 | Depth binary feature pyramid for the detection of small scaled target enhances network |
| CN109583425A (en) * | 2018-12-21 | 2019-04-05 | 西安电子科技大学 | A kind of integrated recognition methods of the remote sensing images ship based on deep learning |
| CN109658447A (en) * | 2018-11-06 | 2019-04-19 | 天津大学 | The nighttime image defogging method kept based on edge details |
| CN109740665A (en) * | 2018-12-29 | 2019-05-10 | 珠海大横琴科技发展有限公司 | Shielded image ship object detection method and system based on expertise constraint |
| CN109902618A (en) * | 2019-02-26 | 2019-06-18 | 青岛海之声科技有限公司 | A kind of sea ship recognition methods and device |
| CN109919113A (en) * | 2019-03-12 | 2019-06-21 | 北京天合睿创科技有限公司 | Ship monitoring method and system and harbour operation prediction technique and system |
| CN110070142A (en) * | 2019-04-29 | 2019-07-30 | 上海大学 | A kind of marine vessel object detection method based on YOLO neural network |
| CN110516565A (en) * | 2019-08-07 | 2019-11-29 | 青岛海之声科技有限公司 | A kind of sea ship recognition methods based on Zernike and SVM |
-
2019
- 2019-12-24 CN CN201911350054.XA patent/CN111695398A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108122003A (en) * | 2017-12-19 | 2018-06-05 | 西北工业大学 | A kind of Weak target recognition methods based on deep neural network |
| CN109472298A (en) * | 2018-10-19 | 2019-03-15 | 天津大学 | Depth binary feature pyramid for the detection of small scaled target enhances network |
| CN109658447A (en) * | 2018-11-06 | 2019-04-19 | 天津大学 | The nighttime image defogging method kept based on edge details |
| CN109583425A (en) * | 2018-12-21 | 2019-04-05 | 西安电子科技大学 | A kind of integrated recognition methods of the remote sensing images ship based on deep learning |
| CN109740665A (en) * | 2018-12-29 | 2019-05-10 | 珠海大横琴科技发展有限公司 | Shielded image ship object detection method and system based on expertise constraint |
| CN109902618A (en) * | 2019-02-26 | 2019-06-18 | 青岛海之声科技有限公司 | A kind of sea ship recognition methods and device |
| CN109919113A (en) * | 2019-03-12 | 2019-06-21 | 北京天合睿创科技有限公司 | Ship monitoring method and system and harbour operation prediction technique and system |
| CN110070142A (en) * | 2019-04-29 | 2019-07-30 | 上海大学 | A kind of marine vessel object detection method based on YOLO neural network |
| CN110516565A (en) * | 2019-08-07 | 2019-11-29 | 青岛海之声科技有限公司 | A kind of sea ship recognition methods based on Zernike and SVM |
Non-Patent Citations (3)
| Title |
|---|
| CHENG Z. YAN ET AL.: "Multiscale saliency detection method for ship targets in synthetic aperture radar images", 《THE JOURNAL OF ENGINEERING》 * |
| 方梦梁 等: "一种光学遥感图像船舶目标检测技术", 《计算机技术与发展》 * |
| 邓超 等: "《数字图像处理与模式识别研究》", 30 June 2018, 北京：地质出版社 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101282A (en) * | 2020-09-25 | 2020-12-18 | 北京瞰天科技有限公司 | Aquatic target identification method and device, electronic equipment and storage medium |
| CN112101282B (en) * | 2020-09-25 | 2024-04-26 | 北京瞰天科技有限公司 | Water target identification method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10943145B2 (en) | Image processing methods and apparatus, and electronic devices | |
| CN110008956B (en) | Invoice key information positioning method, invoice key information positioning device, computer equipment and storage medium | |
| US9014480B2 (en) | Identifying a maximally stable extremal region (MSER) in an image by skipping comparison of pixels in the region | |
| CN111680690B (en) | Character recognition method and device | |
| CN112101386B (en) | Text detection method, device, computer equipment and storage medium | |
| CN111753839A (en) | Text detection method and device | |
| CN110163866A (en) | A kind of image processing method, electronic equipment and computer readable storage medium | |
| WO2023082453A1 (en) | Image processing method and device | |
| CN112329702A (en) | Method and device for rapid face density prediction and face detection, electronic equipment and storage medium | |
| CN111833369A (en) | Alum image processing method, system, medium and electronic device | |
| CN113591719A (en) | Method and device for detecting text with any shape in natural scene and training method | |
| CN110852327A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| CN112801232A (en) | Scanning identification method and system applied to prescription entry | |
| CN115035295A (en) | Remote sensing image semantic segmentation method based on shared convolution kernel and boundary loss function | |
| CN116597466A (en) | Engineering drawing text detection and recognition method and system based on improved YOLOv5s | |
| CN113837965B (en) | Image definition identification method and device, electronic equipment and storage medium | |
| WO2024174726A1 (en) | Handwritten and printed text detection method and device based on deep learning | |
| CN111062347A (en) | Traffic element segmentation method in automatic driving, electronic device and storage medium | |
| CN110929788A (en) | Food material identification method and device, storage medium and electronic equipment | |
| CN110991201A (en) | Bar code detection method and related device | |
| CN111695398A (en) | Small target ship identification method and device and electronic equipment | |
| CN112101323A (en) | Method, system, electronic device and storage medium for identifying title list | |
| CN111582012A (en) | Method and device for detecting small target ship | |
| CN115345895B (en) | Image segmentation method and device for visual detection, computer equipment and medium | |
| Das et al. | Object Detection on Scene Images: A Novel Approach |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200922 |
|
| RJ01 | Rejection of invention patent application after publication |