Search

CN-121540649-B - Method, equipment and medium for improving abundance of ground objects in remote sensing estimated pixels

CN121540649BCN 121540649 BCN121540649 BCN 121540649BCN-121540649-B

Abstract

A method, equipment and medium for improving abundance of ground objects in remote sensing estimated pixels. Belongs to the technical field of remote sensing imaging detection. The method solves the technical problem that the full constraint least square method is affected by the sensor space point spread function, and the obtained ground object abundance has errors. The method comprises the steps of S1, collecting remote sensing image data, S2, estimating the abundance of the ground object in the remote sensing image pixels by adopting an improved full-constraint least square method, introducing the influence of a sensor point spread function and improving a mixed pixel value when estimating the abundance of the ground object in the remote sensing image pixels by adopting the improved full-constraint least square method, and S3, obtaining an estimated result of the abundance of the ground object in the remote sensing image pixels. The method is suitable for the technical field related to remote sensing imaging.

Inventors

  • XU CHAO

Assignees

  • 长春工程学院

Dates

Publication Date
20260512
Application Date
20260122

Claims (6)

  1. 1. The method for improving the abundance of the ground object in the remote sensing estimated pixel is characterized by comprising the following steps of: s1, acquiring remote sensing image data; s2, estimating the abundance of the ground objects in the remote sensing image pixels by adopting an improved full-constraint least square method, introducing the influence of a sensor point spread function when estimating the abundance of the ground objects in the remote sensing image pixels by adopting the improved full-constraint least square method, and improving the mixed pixel value; When the improved full constraint least square method estimates the abundance of ground objects in the remote sensing image pixels, the method comprises the following steps: Obtain no inclusion of True abundance of ground features in the spatial resolution range of (2) By means of Obtaining the real abundance of the ground object in the spatial resolution range Specifically, it will Is inserted into the value of (2) Obtaining ; Wherein, the Represent the first The abundance of the ground object of the species, Representation of And the first The difference value of the spectrum values of the species and the ground object, Representing the original full constraint least square method spectral matrix removal A spectrum matrix remained after the spectrum value of the ground object, , As a matrix of weights, the weight matrix, Representing the original full constraint least square method mixed pixel value and the first The difference value of the spectrum values of the species and the ground object, Representing a transpose operation; S3, obtaining an estimation result of the abundance of the ground object in the remote sensing image pixel.
  2. 2. The method for increasing the abundance of features in a remote sensing estimated pixel of claim 1, The solving process of (1) is specifically as follows: In the original full-constraint least square method, , wherein, Representing the value of the blended picture element, Representing spectral radiation in different wavelength bands, Representing the abundance of the ground object obtained in the original full-constraint least square method, Is a column vector representing white noise in different bands; , Represent the first Spectral values of species.
  3. 3. The method for increasing the abundance of features in a remote sensing estimated pixel of claim 2, Wherein, the method comprises the steps of, Representing an element value of 1 The rank vector of the column, Representing the variance of the random errors contained in each band of the hybrid picture element, Is that The order-unit matrix is used for the data processing, Representing the Hadamard product operation, Value of (2) and The number of rows of (a) is the same.
  4. 4. The method for improving the abundance of features in a remote sensing estimated pixel according to claim 3, Further expressed as , wherein, , , 。
  5. 5. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor implements the steps of the method of any of claims 1-4 when the computer program is executed.
  6. 6. A computer readable storage medium storing computer instructions which, when executed by a processor, implement the steps of the method of any one of claims 1-4.

Description

Method, equipment and medium for improving abundance of ground objects in remote sensing estimated pixels Technical Field The invention belongs to the technical field of remote sensing imaging detection, and particularly relates to a method and a system for improving the abundance of ground features in a remote sensing estimated pixel. Background The image element of the remote sensing image is used as a basic unit for recording surface information, and the interior of the image element often contains complex spectrum mixing phenomenon. Due to the limitation of the spatial resolution of the current remote sensing technology, often, there is not only a single homogeneous ground object in the surface range corresponding to a single pixel. For example, a 30m by 30m Landsat-ETM + pixel may cover a variety of ground types such as vegetation canopy, bare soil, building roof, and shadows. These different features have very different spectral reflectance characteristics and the signal ultimately received by the sensor is effectively a weighted average response of the spectral features of all features within the pixel. This "mixed pixel" phenomenon is widely present in low-to-medium resolution images, and even in high resolution images, spectral mixing inside the pixel is difficult to avoid completely due to the transitivity and complexity of the feature boundary. In the prior art, a full constraint least square method (FCLS) is a common method for estimating the abundance of the ground object in the pixel by remote sensing, but is influenced by a space Point Spread Function (PSF) of a sensor, and the obtained abundance of the ground object has errors. Disclosure of Invention In order to solve the technical problem that the full constraint least square method (FCLS) is influenced by a sensor space Point Spread Function (PSF), and the obtained ground object abundance has errors, the invention provides a method for improving the ground object abundance in a remote sensing estimated pixel. The method comprises the following steps: s1, acquiring remote sensing image data; S2, estimating the abundance of the ground objects in the remote sensing image pixels by adopting an improved full-constraint least square method, introducing the influence of a sensor point spread function when estimating the abundance of the ground objects in the remote sensing image pixels by adopting the improved full-constraint least square method, and improving the mixed pixel value; S3, obtaining an estimation result of the abundance of the ground object in the remote sensing image pixel. Further, when the improved full-constraint least square method estimates the abundance of ground objects in the remote sensing image pixels, the method comprises the following steps: Obtain no inclusion of True abundance of ground features in the spatial resolution range of (2)By means ofObtaining the real abundance of the ground object in the spatial resolution rangeSpecifically, it willIs inserted into the value of (2)Obtaining; Wherein, the Represent the firstThe abundance of the ground object of the species,Representation ofAnd the firstThe difference value of the spectrum values of the species and the ground object,Representing the original full constraint least square method spectral matrix removalA spectrum matrix remained after the spectrum value of the ground object,,As a matrix of weights, the weight matrix,Representing the original full constraint least square method mixed pixel value and the firstThe difference value of the spectrum values of the species and the ground object,Representing the transpose operation. Further, the method comprises the steps of,The solving process of (1) is specifically as follows: In the original full-constraint least square method, , wherein,Representing the value of the blended picture element,Representing spectral radiation in different wavelength bands,Representing the abundance of the ground object obtained in the original full-constraint least square method,Is a column vector representing white noise in different bands;, Represent the first Spectral values of species. Further, the method comprises the steps of,Wherein, the method comprises the steps of,Representing an element value of 1The rank vector of the column,Representing the variance of the random errors contained in each band of the hybrid picture element,Is thatThe order-unit matrix is used for the data processing,Representing the Hadamard product operation,Value of (2) andThe number of rows of (a) is the same. Further, the method comprises the steps of,Further expressed as, wherein,,,。 The method has the beneficial effects that the estimation precision of the traditional full-constraint least square method is lower than expected, the average value of the total error is not more than 0.1 percent by adopting the method, so that the system error cannot be caused, compared with the FCLS method, the improved method reduces the maximum value of the root mean sq