CN112232291A - Ship retrieval method, device, computer equipment and storage medium - Google Patents

Ship retrieval method, device, computer equipment and storage medium Download PDF

Info

Publication number
CN112232291A
CN112232291A CN202011233492.0A CN202011233492A CN112232291A CN 112232291 A CN112232291 A CN 112232291A CN 202011233492 A CN202011233492 A CN 202011233492A CN 112232291 A CN112232291 A CN 112232291A
Authority
CN
China
Prior art keywords
ship
image
feature
target
preset
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
Application number
CN202011233492.0A
Other languages
Chinese (zh)
Inventor
邓练兵
李皓
高妍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhuhai Dahengqin Technology Development Co Ltd
Original Assignee
Zhuhai Dahengqin Technology Development Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Zhuhai Dahengqin Technology Development Co Ltd filed Critical Zhuhai Dahengqin Technology Development Co Ltd
Priority to CN202011233492.0A priority Critical patent/CN112232291A/en
Publication of CN112232291A publication Critical patent/CN112232291A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • 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
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • 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
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Astronomy & Astrophysics (AREA)
  • Remote Sensing (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a ship retrieval method, a ship retrieval device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring an image to be retrieved, and determining whether a ship exists in the image to be retrieved; when the ship exists in the image to be retrieved, cutting a ship image with a target size from the image to be retrieved; carrying out feature extraction on the ship image with the target size to obtain a target ship feature map; and retrieving a target ship picture matched with the target ship feature map from a preset ship feature library. The ship retrieval method reduces the ship retrieval difficulty.

Description

Ship retrieval method, device, computer equipment and storage medium
Technical Field
The present invention relates to the field of image processing, and in particular, to a method and an apparatus for retrieving a ship, a computer device, and a storage medium.
Background
The motion condition of the ship is an important target of the perception information of the coastal area, and is an important content for ship detection in the field of remote sensing images. Since the satellite images of the coastal areas contain a plurality of different scenes, including building areas, vegetation areas, mountains, rivers and the like, ships occupy a small percentage of the scenes. The ship is used as a small ground object, has relatively less characteristic information, and has certain difficulty in searching the ship.
Disclosure of Invention
The embodiment of the invention provides a ship retrieval method, a ship retrieval device, computer equipment and a storage medium, which are used for reducing the difficulty of ship retrieval.
The invention provides a ship retrieval method, which comprises the following steps:
acquiring an image to be retrieved, and determining whether a ship exists in the image to be retrieved;
when determining that ships exist in the image to be retrieved, cutting a ship image with a target size from the image to be retrieved;
carrying out feature extraction on the ship image with the target size to obtain a target ship feature map;
and retrieving a target ship picture matched with the target ship feature map from a preset ship feature library.
Preferably, retrieving a ship picture matched with the target ship feature map from a preset ship feature library, including:
extracting a plurality of preset ship feature graphs from a ship feature library;
calculating the Euclidean distance between the target ship feature map and each preset ship feature map:
Figure BDA0002765977700000021
wherein L is Euclidean distance, x and y respectively represent a target characteristic diagram and a preset ship characteristic diagram, i represents the ith as a characteristic vector, and n represents the characteristic vector dimension of an image;
and taking the ship picture corresponding to the minimum Euclidean distance preset ship feature graph as a target ship picture.
Preferably, the target ship feature map and the preset ship feature map both contain GLCM features, or the target ship feature map and the preset ship feature map both contain SIFT features.
Preferably, the preset ship feature library is generated by the following steps:
acquiring a preset ship picture set, wherein the preset ship picture set comprises a plurality of ship pictures;
removing interference of each ship picture;
and performing feature extraction on each ship picture subjected to interference removal by adopting a preset feature extraction model, and storing the extracted feature in a database to obtain a preset ship feature library.
Preferably, the feature extraction model comprises 13 convolutional layers in total of 5 groups and 3 fully-connected layers, wherein each of the first 2 convolutional layers comprises 2 convolutional layers, each of the last 3 convolutional layers comprises 3 convolutional layers, and one maximum pooling layer is added after each convolutional layer, and the size of a convolutional core of each convolutional layer is 3 × 3.
Preferably, after calculating each euclidean distance, the ship retrieval method further includes: all euclidean distances are sorted.
The embodiment of the present invention further provides a ship retrieval device, which is characterized by comprising:
the acquisition unit is used for acquiring the image to be retrieved and determining whether a ship exists in the image to be retrieved;
the cutting unit is used for cutting a ship image with a target size from the image to be retrieved when the ship exists in the image to be retrieved;
the characteristic extraction unit is used for extracting the characteristics of the ship image with the target size to obtain a target ship characteristic diagram;
and the retrieval unit is used for retrieving the target ship picture matched with the target ship feature picture from a preset ship feature library.
Preferably, the retrieval unit includes:
the extracting subunit is used for extracting a plurality of preset ship feature maps from the ship feature library;
the calculating subunit is used for calculating the Euclidean distance between the target ship feature map and each preset ship feature map:
Figure BDA0002765977700000031
wherein L is Euclidean distance, x and y respectively represent a target characteristic diagram and a preset ship characteristic diagram, i represents the ith as a characteristic vector, and n represents the characteristic vector dimension of an image;
and the result determining subunit is used for taking the ship picture corresponding to the preset ship characteristic graph with the minimum Euclidean distance as the target ship picture.
The embodiment of the invention also provides computer equipment which comprises a memory and a processor and is characterized in that a ship retrieval program is stored in the memory, and the processor is used for realizing the steps of the ship retrieval method when executing the ship retrieval program.
The embodiment of the invention also provides a computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and the computer program is executed by a processor to realize the steps of the ship retrieval method.
According to the ship retrieval method, the ship retrieval device, the computer equipment and the storage medium, after the image to be retrieved is obtained, ship identification is carried out on the image to be retrieved, when the ship exists in the image to be retrieved, the image to be retrieved is cut, so that the background is removed to the maximum extent, then the cut ship image is subjected to feature extraction, and the extracted feature image is matched with the target ship feature image in the ship feature library, so that the effect of retrieving the ship image matched with the extracted feature image according to the existing image to be retrieved is achieved. In addition, due to the fact that the image to be retrieved is cut, the influence of background interference on the accuracy of ship retrieval is avoided, background features are reduced to the maximum extent, more ship features are reserved, and the retrieval difficulty is reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments of the present invention will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.
FIG. 1 is a flow chart of a ship retrieval method in one embodiment of the invention;
FIG. 2 is a flow chart of a ship retrieval method in another embodiment of the present invention;
fig. 3 is a schematic block diagram of a ship retrieval apparatus according to an embodiment of the present invention.
Detailed Description
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, 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 ship retrieval method provided by the embodiment of the invention can be used for retrieving the target ship image matched with the ship image to be retrieved from the massive ship images. Specifically, as shown in fig. 1, the method includes the steps of:
s10: and acquiring an image to be retrieved, and determining whether a ship exists in the image to be retrieved.
The ship image to be retrieved may be an image of a ship moving on the sea area captured by a specific camera, and the ship image to be retrieved may be a remote sensing image, and the ship image to be retrieved is not particularly limited. The above determining whether the ship sending process exists in the image to be retrieved can also be regarded as a ship positioning process, and specifically, the ship positioning can be performed by adopting a Selective Search algorithm.
S20: and cutting out the ship image with the target size from the image to be retrieved when the ship exists in the image to be retrieved.
Because the ship usually only occupies a small part of the image in the image to be retrieved, a large amount of background redundancy exists in images of different ships, and meanwhile, because the similarity of the sea surface background has a great influence on the retrieval precision of subsequent ships, the ship image to be retrieved is cut to cut the redundant partial image, and the ship partial image is kept as much as possible.
S30: and carrying out feature extraction on the ship image with the target size to obtain a target ship feature map.
Illustratively, the step can be that ship images with target sizes are input into a convolutional neural network which is trained in advance, convolutional layer features and full-connected layer features of ships are extracted in the process of forward propagation of the convolutional neural network, and then the convolutional layer features and the full-connected layer features are combined to obtain a target ship feature map.
The target ship feature map in this step may be a feature map including GLCM (Gray Level Co-objective Matrix) features, or may be a feature map including SIFT features.
The Gray Level Co-oven Matrix (GLCM) reflects the spatial dependency of the Gray Level in the texture of the image, and is a texture feature analysis method with wide application. Assuming that I (x, y) represents a grayscale image, P is any region of the image, and Q is the set of pixel pairs in the region that satisfy a particular spatial relationship, the corresponding normalized grayscale co-occurrence matrix can be expressed as:
Figure BDA0002765977700000061
wherein x is2＝x1+dcosθ,y2＝y1+ dsin θ, card (Q) denotes the number of pixel pairs satisfying the Q-set condition, and d and θ denote the pixel pitch and direction.
In practical use, in order to reduce the amount of calculation, it is generally necessary to quantize an image to reduce the number of gray levels, and θ generally takes four directions of 0, 45 °, 90 °, and 135 °. On the basis of the gray level co-occurrence matrix, a plurality of statistics can be calculated to describe the texture features of the image, and the commonly used statistics include the following four types:
angular second moment (energy):
Figure BDA0002765977700000062
entropy:
Figure BDA0002765977700000063
correlation:
Figure BDA0002765977700000064
contrast ratio:
Figure BDA0002765977700000065
wherein
Figure BDA0002765977700000066
Figure BDA0002765977700000067
(i,j)∈[1,Ng]Is the value of an element in the gray level co-occurrence matrix, NgRepresenting the number of gray levels of the image;
when extracting the SIFT feature points of the image, it is necessary to divide the neighborhood around all the key points into 4 × 4 sub-regions, describe the key points in the gradient strength information of 8 directions of each sub-region seed point, and finally obtain a 128-dimensional feature vector. Assuming that i represents any one image in the image library, the SIFT feature point X of each imageiCan be represented by the following formula:
Figure BDA0002765977700000071
wherein the content of the first and second substances,
Figure BDA0002765977700000072
representing a 128-dimensional SIFT feature point vector, representingThe number of SIFT feature points of the image.
S40: and retrieving a target ship picture matched with the target ship feature map from a preset ship feature library.
In this step, the ship feature library is a feature library obtained by storing a preset ship feature map obtained by extracting features of ship pictures in a preset ship picture set in a database. Specifically, the preset ship feature library may be generated by:
s41: acquiring a preset ship picture set, wherein the preset ship picture set comprises a plurality of ship pictures;
s42: removing interference of each ship picture;
because the ship picture contains more objects or backgrounds, in order to remove background interference and improve the ship retrieval speed, a ship target in the picture can be identified by using a target detection method.
S43: and performing feature extraction on each ship picture subjected to interference removal by adopting a preset feature extraction model, and storing the extracted feature in a database to obtain a preset ship feature library.
In this step, the initial model of the feature extraction model may be VGG-16, which has 16 layers including 13 convolutional layers and 3 fully-connected layers. Similar to the 5-layer structure of AlexNet, the 13 convolutional layers of VGG-16 are divided into 5 groups, with one max-pooling layer added after each group of convolutional layers. The 5 convolutional layers all use a 3 x 3 convolutional kernel, the first two groups of convolutional layers each contain 2 convolutional layers, and the last three groups of convolutional layers each contain 3 convolutional layers. The VGG-16 network improves the performance of the network by deepening the number of layers in the network and does not increase the computational overhead because a smaller convolution kernel is used.
It will be appreciated that in order to facilitate feature matching and improve the accuracy of the search, the preset ship feature maps in the ship feature library should contain features that are consistent with those contained in the target ship feature map, for example, if the target ship feature map contains GLCM features, the feature maps in the ship feature library should also contain GLCM features.
In an embodiment where both the target ship feature map and the preset ship feature map contain GLCM features, before specific retrieval, the mean and standard deviation of the above four statistics (four directions) may be calculated respectively and concatenated to obtain an 8-dimensional texture feature vector, and then feature matching is performed.
When specific retrieval (feature matching) is carried out, an exhaustion method can be adopted, a plurality of preset ship feature graphs are extracted from a ship feature library, then the target ship feature graph is matched with each preset ship feature graph one by one, and the target ship picture is determined according to a matching result. Specifically, as shown in fig. 2, the process may include the steps of:
s44: extracting a plurality of preset ship feature graphs from a ship feature library;
s45: calculating the Euclidean distance between the target ship feature map and each preset ship feature map:
Figure BDA0002765977700000081
wherein L is Euclidean distance, x and y respectively represent a target characteristic diagram and a preset ship characteristic diagram, i represents the ith as a characteristic vector, and n represents the characteristic vector dimension of an image;
s46: and taking the ship picture corresponding to the minimum Euclidean distance preset ship feature graph as a target ship picture.
It should be noted that after the euclidean distance between the target ship feature map and each preset ship feature map is calculated, the euclidean distances are also required to be sorted to determine the preset ship feature map with the minimum euclidean distance, and the ship picture corresponding to the preset ship feature map with the minimum euclidean distance is taken as the target ship picture.
In the embodiment, after the image to be retrieved is obtained, the image to be retrieved is subjected to ship identification, when the ship exists in the image to be retrieved, the image to be retrieved is cut to remove the background to the maximum extent, then the cut ship image is subjected to feature extraction, and the extracted feature map is matched with the target ship feature map in the ship feature library, so that the effect of retrieving the ship image matched with the extracted feature map according to the existing image to be retrieved is realized. In addition, due to the fact that the image to be retrieved is cut, the influence of background interference on the accuracy of ship retrieval is avoided, background features are reduced to the maximum extent, more ship features are reserved, and the retrieval difficulty is reduced.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
In one embodiment, a ship retrieval device is provided, and the ship retrieval device corresponds to the ship retrieval method in the above embodiments one to one. As shown in fig. 3, the ship retrieval apparatus includes:
the acquiring unit 10 is used for acquiring an image to be retrieved and determining whether a ship exists in the image to be retrieved;
a cutting unit 20 for cutting out a ship image of a target size from the image to be retrieved when it is determined that the ship exists in the image to be retrieved;
the characteristic extraction unit 30 is used for carrying out characteristic extraction on the ship image with the target size to obtain a target ship characteristic diagram;
and the retrieval unit 40 is used for retrieving the target ship picture matched with the target ship feature map from a preset ship feature library.
Preferably, the retrieval unit 40 includes:
an extracting subunit 41, configured to extract a plurality of preset ship feature maps from the ship feature library;
a calculating subunit 42, configured to calculate a euclidean distance between the target ship feature map and each preset ship feature map:
Figure BDA0002765977700000101
wherein L is Euclidean distance, x and y respectively represent a target characteristic diagram and a preset ship characteristic diagram, i represents the ith as a characteristic vector, and n represents the characteristic vector dimension of an image;
and the result determining subunit 43 is configured to use the ship picture corresponding to the preset ship feature map with the minimum euclidean distance as the target ship picture.
For specific limitations of the ship retrieval device, reference may be made to the above limitations of the ship retrieval method, which are not described herein again. The various modules in the ship retrieval device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing a preset ship characteristic map. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a ship retrieval method.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which computer program, when being executed by a processor, realizes the above-mentioned vessel retrieval method.
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, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules, so as to perform all or part of the functions described above.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.

Claims (10)

1. A method for ship retrieval, comprising:
acquiring an image to be retrieved, and determining whether a ship exists in the image to be retrieved;
when the ship exists in the image to be retrieved, cutting a ship image with a target size from the image to be retrieved;
carrying out feature extraction on the ship image with the target size to obtain a target ship feature map;
and retrieving a target ship picture matched with the target ship feature map from a preset ship feature library.
2. The ship retrieval method of claim 1, wherein the retrieving of the ship picture matching the target ship feature map from a preset ship feature library comprises:
extracting a plurality of preset ship feature graphs from the ship feature library;
calculating Euclidean distance between the target ship feature map and each preset ship feature map:
Figure FDA0002765977690000011
wherein L is the Euclidean distance, x and y respectively represent the target feature map and the preset ship feature map, i represents the ith as a feature vector, and n represents the feature vector dimension of an image;
and taking the ship picture corresponding to the minimum Euclidean distance preset ship feature graph as the target ship picture.
3. The vessel retrieval method of claim 2, wherein the target vessel feature map and the preset vessel feature map both contain GLCM features, or wherein the target vessel feature map and the preset vessel feature map both contain SIFT features.
4. The ship retrieval method of claim 1, wherein the preset ship feature library is generated by:
acquiring a preset ship picture set, wherein the preset ship picture set comprises a plurality of ship pictures;
removing interference from each ship picture;
and performing feature extraction on each ship picture subjected to interference removal by adopting a preset feature extraction model, and storing the ship pictures in a database to obtain the preset ship feature library.
5. The vessel search method of claim 4, wherein said feature extraction model comprises a total of 13 convolutional layers of 5 sets and 3 fully-connected layers, wherein each of the first 2 convolutional layers comprises 2 convolutional layers, each of the last 3 convolutional layers comprises 3 convolutional layers, and a maximum pooling layer is added after each convolutional layer, and the convolutional kernel size of each convolutional layer is 3 x 3.
6. The vessel retrieval method of claim 2, wherein after each of said euclidean distances is calculated, said vessel retrieval method further comprises: sorting all the Euclidean distances.
7. A ship retrieval apparatus, comprising:
the system comprises an acquisition unit, a retrieval unit and a retrieval unit, wherein the acquisition unit is used for acquiring an image to be retrieved and determining whether a ship exists in the image to be retrieved;
the cutting unit is used for cutting a ship image with a target size from the image to be retrieved when the ship exists in the image to be retrieved;
the characteristic extraction unit is used for extracting the characteristics of the ship image with the target size to obtain a target ship characteristic diagram;
and the retrieval unit is used for retrieving the target ship picture matched with the target ship feature map from a preset ship feature library.
8. The ship retrieval device of claim 6, wherein the retrieval unit comprises:
the extracting subunit is used for extracting a plurality of preset ship feature maps from the ship feature library;
a calculating subunit, configured to calculate a euclidean distance between the target ship feature map and each of the preset ship feature maps:
Figure FDA0002765977690000031
wherein L is the Euclidean distance, x and y respectively represent the target feature map and the preset ship feature map, i represents the ith as a feature vector, and n represents the feature vector dimension of an image;
and the result determining subunit is used for taking the ship picture corresponding to the preset ship feature map with the minimum Euclidean distance as the target ship picture.
9. A computer arrangement comprising a memory and a processor, characterized in that said memory has stored therein a ship retrieval program, said processor being adapted to carry out the steps of the ship retrieval method according to any of claims 1 to 6 when executing said ship retrieval program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the vessel retrieval method according to any one of claims 1 to 5.
CN202011233492.0A 2020-11-06 2020-11-06 Ship retrieval method, device, computer equipment and storage medium Pending CN112232291A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011233492.0A CN112232291A (en) 2020-11-06 2020-11-06 Ship retrieval method, device, computer equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011233492.0A CN112232291A (en) 2020-11-06 2020-11-06 Ship retrieval method, device, computer equipment and storage medium

Publications (1)

Publication Number Publication Date
CN112232291A true CN112232291A (en) 2021-01-15

Family

ID=74122095

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011233492.0A Pending CN112232291A (en) 2020-11-06 2020-11-06 Ship retrieval method, device, computer equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112232291A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN109977897A (en) * 2019-04-03 2019-07-05 厦门兴康信科技股份有限公司 A kind of ship's particulars based on deep learning recognition methods, application method and system again
CN110287350A (en) * 2019-06-29 2019-09-27 北京字节跳动网络技术有限公司 Image search method, device and electronic equipment
CN111553182A (en) * 2019-12-26 2020-08-18 珠海大横琴科技发展有限公司 Ship retrieval method and device and electronic equipment
CN111695572A (en) * 2019-12-27 2020-09-22 珠海大横琴科技发展有限公司 Ship retrieval method and device based on convolutional layer feature extraction

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN109977897A (en) * 2019-04-03 2019-07-05 厦门兴康信科技股份有限公司 A kind of ship's particulars based on deep learning recognition methods, application method and system again
CN110287350A (en) * 2019-06-29 2019-09-27 北京字节跳动网络技术有限公司 Image search method, device and electronic equipment
CN111553182A (en) * 2019-12-26 2020-08-18 珠海大横琴科技发展有限公司 Ship retrieval method and device and electronic equipment
CN111695572A (en) * 2019-12-27 2020-09-22 珠海大横琴科技发展有限公司 Ship retrieval method and device based on convolutional layer feature extraction

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
王彦情;马雷;田原;: "光学遥感图像舰船目标检测与识别综述", 自动化学报, no. 09, pages 1029 - 1039 *
王鹏，曹聚亮，胡小平: "水下地磁导航适配性研究", 28 February 2017, 国防工业出版社, pages: 96 *

Similar Documents

Publication Publication Date Title
CN110738207B (en) Character detection method for fusing character area edge information in character image
CN110751134B (en) Target detection method, target detection device, storage medium and computer equipment
US10269147B2 (en) Real-time camera position estimation with drift mitigation in incremental structure from motion
CN108960230B (en) Lightweight target identification method and device based on rotating rectangular frame
US20180315232A1 (en) Real-time incremental 3d reconstruction of sensor data
US20180315222A1 (en) Real-time image undistortion for incremental 3d reconstruction
CN111291637A (en) Face detection method, device and equipment based on convolutional neural network
CN111191533B (en) Pedestrian re-recognition processing method, device, computer equipment and storage medium
CN111753669A (en) Hand data identification method, system and storage medium based on graph convolution network
CN114092833A (en) Remote sensing image classification method and device, computer equipment and storage medium
CN114241377A (en) Ship target detection method, device, equipment and medium based on improved YOLOX
CN114419570A (en) Point cloud data identification method and device, electronic equipment and storage medium
CN111179270A (en) Image co-segmentation method and device based on attention mechanism
CN111373393B (en) Image retrieval method and device and image library generation method and device
CN112613553A (en) Picture sample set generation method and device, computer equipment and storage medium
CN114359665A (en) Training method and device of full-task face recognition model and face recognition method
CN113095310B (en) Face position detection method, electronic device and storage medium
CN111666931A (en) Character and image recognition method, device and equipment based on mixed convolution and storage medium
CN115272691A (en) Training method, recognition method and equipment for steel bar binding state detection model
CN113704276A (en) Map updating method and device, electronic equipment and computer readable storage medium
CN111582013A (en) Ship retrieval method and device based on gray level co-occurrence matrix characteristics
CN112232291A (en) Ship retrieval method, device, computer equipment and storage medium
CN113077475B (en) Visual positioning method, device, system, mobile robot and storage medium
CN111178202B (en) Target detection method, device, computer equipment and storage medium
CN112907662A (en) Feature extraction method and device, electronic equipment and storage medium

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