A Curated Image Parameter Dataset from Solar Dynamics Observatory Mission

This API is the product of our study abstracted below:

We provide a massive image parameter dataset extracted from the Solar Dynamics Observatory (SDO) mission’s AIA instrument, for the period of January 2011 through the current date, with the cadence of six minutes, for nine wavelength channels. Towards better results in the region classification of solar events, in this work, we improve upon the performance of a set of ten image parameters when utilized in the task of classifying regions of the AIA images. This is accomplished through an in depth analysis of various assumptions made in the calculation of these parameters, in order to find areas of improving the outcome of the stated classification task. Then, where possible, a method for finding an appropriate settings for the parameter calculations was devised, as well as a validation task that is used to show our improved results. This process is repeated for each of the nine different wavelength channels that are included in our analysis. In addition to investigating the effects of assumptions made during the calculation process, we also include comparisons of JP2 and FITS image formats in a pixel-based, supervised classification task, by tuning the parameters specific to the format of the images from which they are extracted. The results of these comparisons show that utilizing JP2 images, which are significantly smaller files, is not detrimental to the region classification task that these parameters were originally intended for. Finally, we compute the tuned parameters on the AIA images and to make the resultant dataset easily accessible for others, we provide this public API for random access to the calculated parameters.

1 ) Web API in Python

It's quite easy to use the Python API to get data, relying on our random-access web API. See the examples below:

									
                                                    from datetime import datetime
                                                    from constants.CONSTANTS import *
                                                    from aia_image_api import imageparam_getter as ipg
                                                    dt = datetime.strptime('2012-02-13T20:10:00', '%Y-%m-%dT%H:%M:%S')
                                                    aia_wave = AIA_WAVE.AIA_171
                                                    image_size = IMAGE_SIZE.P2000
                                                    param_id = IMAGE_PARAM.ENTROPY

                                                    # Example 1: Retrieve an AIA image:
                                                    img = ipg.get_aia_image_jpeg(dt, aia_wave, image_size)
                                                    img.show()

                                                    # Example 2: Retrieve a heatmap of the computed image parameters:
                                                    heatmap = ipg.get_aia_imageparam_jpeg(dt, aia_wave, image_size, param_id)
                                                    heatmap.show()

                                                    # Example 3: Retrieve an xml file of the computed image parameters, and convert it to 'ndarray':
                                                    xml = ipg.get_aia_imageparam_xml(dt, aia_wave)
                                                    res = ipg.convert_param_xml_to_ndarray(xml)
                                                    print(res.reshape(64 * 64 * 10).tolist())

For more details, visit the repoditory at https://bitbucket.org/gsudmlab/imageparams_api/src/master/.

2) Web API

Or, you can access the data directly through the Web API by sending the following requests:

AIA Images

Parameter Images

Image Parameter One Image

Option Descriptions:

  • wave: Wavelength filter of AIA image from SDO
  • starttime: Start time of the query, which takes values in the form of "2012-01-22T00:00:00"

Image Parameter Range of Images

Option Descriptions:

  • wave: Wavelength filter of AIA image from SDO
  • starttime: Start time of the query, which takes values in the form of "2012-01-22T00:00:00"
  • endtime: End time of the query, which takes values in the form of "2012-02-22T00:00:00"
  • limit: (optional: default/maximum=100) Denotes the maximum number of images that the query will return in a single request.
  • offset: (optional: default=0) The offset in number from the start of the list of results that meet the requirements passed into the method. Used to perform paging when used in conjunction with limit.
  • step: (optional: default=1) The sample rate of individual images from the dataset. The parameter data is calculated at a 6 minute cadence rate, but users may not wish to utilize the full cadence rate. This variable gives them the ability to take every Nth image from the dataset.

Image Header Information (meta data)

Using these requests, a small subset of the header information of each AIA image can be retrieved. This is needed for integration of spatial information of solar events, and also it allows exclusion of "bad" images, images which are too noisy, flipped, captured during eclipse, or unaligned, and in general, scientifically unreliable. The response contains the following pieces of information:

  • X0: X Coordinate of the center of Sun's disk in 4096X4096-pixel images, in pixels. [Integer]
  • Y0: Y Coordinate of the center of Sun's disk in 4096X4096-pixel images, in pixels. [Integer]
  • DSUN: Distance from Sun's center to SDO in meter. [Double]
  • R_SUN: Radius of Sun in the image. [Double]
  • CDELT: Pixel spacing per index value along image axes.
  • QUALITY: A 32-bit binary number, whose each bit represents one quality issue.
For more details, see AIA-SDO_FITS_Keyword_Document.pdf

Parameters & Wavelength Channels

The available wavelength channels and image parameters are listed here:

'wave': {94, 131, 171, 193, 211, 304, 355, 1600, 1700}
'param': {1: Entropy, 2: Mean, 3: Standard Deviation, 4: Fractal Dimension, 5: Skewness, 6: Kurtosis, 7: Uniformity, 8: Relative Smoothness, 9: Tamura Contrast, 10: Tamura Directionality}

Heatmap of Image Parameters on an AIA Image (171 Å)

Heliophysics Events Knowledge Base Clone

This API is the product of our study abstracted below:

Over the last decade, the volume of solar big data have increased immensely. However, the availability and standardization of solar data resources has not received much attention primarily due to the scattered structure among different data providers, lack of consensus on data formats and querying capabilities on metadata. Moreover, there is limited access to the derived solar data such as image parameters extracted either from solar images or tracked solar events. In this paper, we introduce the Integrated Solar Database (ISD), which aims to integrate the heterogeneous solar data sources. In ISD, we store solar event metadata, tracked and interpolated solar events. ISD offers spatiotemporal and aggregate queries served via a web Application Program Interface (API) and visualized through a web interface.

Temporal and SpatioTemporal Event Query