阅读说明：本技术 一种基于psvm和crnn的遥感影像分类显示装置 (Remote sensing image classification display device based on PSVM and CRNN ) 是由 俞颖 黄风华 林小彬 于 2019-09-06 设计创作，主要内容包括：本发明公开了一种基于PSVM和CRNN的遥感影像分类显示装置,包括摄像采集模块,所述摄像采集模块连接有摄像机,所述摄像采集模块分别电性连接有第一处理单元和第二处理单元,所述第一处理单元和第二处理单元电性连接有数据合并单元,所述数据合并单元电性连接有结果反馈单元,本发明具有第一处理单元和第二处理单元,基于混合式并行支撑向量机技术,引入协作式递归神经网络来进行遥感影像的分类或识别,将多类分类问题分解成两个约束优化问题,引入CRNN进行支撑向量机的参数学习可以获得问题的全局最优解,与此同时,采用PSVM技术可以有效地降低解题的规模,提高分类速度,从而获得良好的分类效果。(The invention discloses a remote sensing image classification display device based on PSVM and CRNN, which comprises a camera acquisition module, wherein the camera acquisition module is connected with a camera, the camera acquisition module is respectively and electrically connected with a first processing unit and a second processing unit, the first processing unit and the second processing unit are electrically connected with a data merging unit, and the data merging unit is electrically connected with a result feedback unit. Thereby obtaining good classification effect.)

1. The utility model provides a remote sensing image classification display device based on PSVM and CRNN, includes collection module (2) of making a video recording, its characterized in that: the camera shooting acquisition module (2) is connected with a camera (1), the camera shooting acquisition module (2) is respectively and electrically connected with a first processing unit (3) and a second processing unit (4), the first processing unit (3) and the second processing unit (4) are electrically connected with a data merging unit (5), the data merging unit (5) is electrically connected with a result feedback unit (6), one end of the result feedback unit (6) is electrically connected with a local storage device (10), the local storage device (10) is connected with a display (12) through an I/O interface (11), the other end of the result feedback unit (6) is electrically connected with a network adapter (7), the network adapter (7) is electrically connected with a communication module (8), and the communication module (8) is connected with a cloud storage device (9).

2. The PSVM and CRNN-based remote sensing image classification display device according to claim 1, wherein: the first processing unit (3) comprises a first ARM chip (31), a first RAM memory (32) and a first ROM memory (33).

3. The PSVM and CRNN-based remote sensing image classification display device according to claim 1, wherein: the second processing unit (4) comprises a second ARM chip (41), a second RAM memory (42) and a second ROM memory (43).

4. The PSVM and CRNN-based remote sensing image classification display device according to claim 1, wherein: the communication module (8) is a 3G, 4G or WIFI module.

5. The PSVM and CRNN-based remote sensing image classification display device according to claim 1, wherein: the I/O interface (11) is a parallel interface, a serial interface, a USB interface or an optical fiber interface.

6. The PSVM and CRNN-based remote sensing image classification display device according to claim 1, wherein: the local storage device (10) is a raspberry pi RS-3B.

Technical Field

The invention relates to the technical field of image classification devices, in particular to a remote sensing image classification display device based on PSVM and CRNN.

Background

The remote sensing image classification refers to a process of dividing each pixel in the remote sensing image into different surface feature classes. The classification basis mainly comprises surface feature spectral characteristics, surface feature shape characteristics, spatial relationship characteristics and the like, and most of the current researches still classify the remote sensing images based on the surface feature spectral characteristics.

The traditional classification method based on the pixels takes a single pixel as a unit, focuses on local parts and ignores the geometrical structure condition of a whole image spot nearby, so that the phenomenon of salt and pepper is easily generated, the accuracy of characteristic information extraction is restricted, however, the existing technology cannot accurately classify, and therefore, a remote sensing image classification display device based on PSVM and CRNN is provided.

Disclosure of Invention

The invention aims to provide a remote sensing image classification display device based on PSVM and CRNN, which aims to solve the problems in the background technology.

In order to achieve the purpose, the invention provides the following technical scheme: the utility model provides a remote sensing image classification display device based on PSVM and CRNN, includes the collection module of making a video recording, the collection module of making a video recording is connected with the camera, the collection module of making a video recording electric connection has first processing unit and second processing unit respectively, first processing unit and second processing unit electric connection have data merging unit, data merging unit electric connection has result feedback unit, result feedback unit's one end electric connection has local storage device, local storage device has the display through IO interface connection, result feedback unit's other end electric connection has network adapter, network adapter electric connection has communication module, communication module is connected with cloud storage device.

The invention aims to provide a method for classifying and identifying remote sensing images by introducing a collaborative recurrent neural network based on a hybrid parallel support vector machine technology, decomposing a multi-class classification problem into two constraint optimization problems, and introducing CRNN to learn parameters of a support vector machine to obtain a global optimal solution of the problem.

Preferably, the first processing unit includes a first ARM chip, a first RAM memory, and a first ROM memory.

Preferably, the second processing unit includes a second ARM chip, a second RAM memory, and a second ROM memory.

Preferably, the communication module is a 3G, 4G or WIFI module.

Preferably, the I/O interface is a parallel interface, a serial interface, a USB interface, or an optical fiber interface.

Preferably, the local storage device is raspberry pi RS-3B.

Compared with the prior art, the invention has the beneficial effects that: the method is provided with a first processing unit and a second processing unit, a collaborative recurrent neural network is introduced to classify or identify the remote sensing images based on the hybrid parallel support vector machine technology, the multi-class classification problem is decomposed into two constraint optimization problems, and the CRNN is introduced to learn the parameters of the support vector machine to obtain the global optimal solution of the problem.

Drawings

FIG. 1 is a schematic structural view of the present invention;

FIG. 2 is a schematic diagram of a first processing unit according to the present invention;

FIG. 3 is a diagram illustrating a second exemplary processing unit according to the present invention.

In the figure: 1. a camera; 2. a camera shooting and collecting module; 3. a first processing unit; 31. a first ARM chip; 32. a first RAM memory; 33. a first ROM memory; 4. a second processing unit; 41. a second ARM chip; 42. a second RAM memory; 43. a second ROM memory; 5. a data merging unit; 6. a result feedback unit; 7. a network adapter; 8. a communication module; 9. a cloud storage device; 10. a local storage device; 11. an I/O interface; 12. a display.

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 only a part of the embodiments of the present invention, and not all of the embodiments. 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.

Referring to fig. 1-3, the present invention provides a technical solution: the utility model provides a remote sensing image classification display device based on PSVM and CRNN, includes collection module 2 of making a video recording, collection module 2 of making a video recording is connected with camera 1, collection module 2 of making a video recording electric connection respectively has first processing unit 3 and second processing unit 4, first processing unit 3 and 4 electric connection of second processing unit have data merging unit 5, 5 electric connection of data merging unit has result feedback unit 6, the one end electric connection of result feedback unit 6 has local storage device 10, local storage device 10 is connected with display 12 through IO interface 11, the other end electric connection of result feedback unit 6 has network adapter 7, network adapter 7 electric connection has communication module 8, communication module 8 is connected with cloud storage device 9.

Specifically, the first processing unit 3 includes a first ARM chip 31, a first RAM memory 32, and a first ROM memory 33.

Specifically, the second processing unit 4 includes a second ARM chip 41, a second RAM memory 42, and a second ROM memory 43.

Specifically, the communication module 8 is a 4G module.

Specifically, the I/O interface 11 is a parallel interface.

Specifically, the local storage device 10 is a raspberry pi RS-3B.

Specifically, when the system is used, the system is provided with a first processing unit 3 and a second processing unit 4, a collaborative recurrent neural network is introduced to classify or identify remote sensing images based on a hybrid parallel support vector machine technology, a multi-class classification problem is decomposed into two constraint optimization problems, a CRNN is introduced to learn parameters of a support vector machine, so that a global optimal solution of the problems can be obtained, meanwhile, the PSVM technology can be adopted to effectively reduce the scale of the solution problems, improve the classification speed, and further obtain a good classification effect.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.