BEGIN:VCALENDAR VERSION:2.0 PRODID:-//pretalx//talks.staging.osgeo.org//foss4g-europe-2024-academic-tra ck//talk//DQCQMS BEGIN:VTIMEZONE TZID:EET BEGIN:STANDARD DTSTART:20000101T000000 RRULE:FREQ=YEARLY;BYMONTH=1;UNTIL=20001231T220000Z TZNAME:EET TZOFFSETFROM:+0200 TZOFFSETTO:+0200 END:STANDARD BEGIN:STANDARD DTSTART:20021027T050000 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10 TZNAME:EET TZOFFSETFROM:+0300 TZOFFSETTO:+0200 END:STANDARD BEGIN:DAYLIGHT DTSTART:20020331T040000 RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3 TZNAME:EEST TZOFFSETFROM:+0200 TZOFFSETTO:+0300 END:DAYLIGHT END:VTIMEZONE BEGIN:VEVENT UID:pretalx-foss4g-europe-2024-academic-track-DQCQMS@talks.staging.osgeo.or g DTSTART;TZID=EET:20240704T150000 DTEND;TZID=EET:20240704T153000 DESCRIPTION:Spatial pattern is an inherent property visible in many spatial variables. Spatial patterns are often at the heart of many geographical s tudies\, where we search for existing hot spots\, correlations\, and outli ers. They may be exhibited in various forms\, depending on the type of dat a and the underlying processes that generated the data. Here\, we will foc us on spatial patterns in spatial rasters\, but the concept can be extende d to other types of spatial data\, including vector data and point clouds. \n\nPatterns in spatial raster data may have many forms. We may think of s patial patterns for continuous rasters as an interplay between intensity a nd spatial autocorrelation (e.g.\, elevation) or between composition and c onfiguration for categorical rasters (e.g.\, land cover) (Gustafson\, 1998 ). Intensity relates to the range and distribution of values of a given va riable\, while spatial autocorrelation is a tendency for nearby values of a given variable to be more similar than those that are further apart. On the other hand\, composition is the number of cells belonging to each map category\, while configuration represents their spatial arrangement. Ano ther distinction is between the data dimensionality. The most common situa tion is when we only use one layer of given data (e.g.\, an elevation map or a land cover product for one year). However\, we may also be interested in sets of variables (layers\, bands)\, such as hyperspectral data\, time series\, or proportions of classes. An additional special case is the RGB representation of the data.\n\nAssessing the similarity of spatial patter ns is a common task in many fields\, including remote sensing\, ecology\, and geology. This procedure may encapsulate many types of comparisons: com paring the same variable(s) for different areas\, comparing different data sets (e.g.\, different sensors)\, or comparing the same area but at differ ent times.\n\nGiven various possible scientific questions and the fact tha t we have a plethora of forms of spatial data\, there is no universal meth od for assessing similarity between two spatial patterns. The basic method is a visual inspection\; however\, it is highly subjective\, both from th e observer’s and visualization type’s perspectives. Fairly straightfor ward other approaches are to create a difference map\, count changed pixel s\, or look at the distribution of the values. More advanced methods inclu de the use of machine learning algorithms. However\, these methods are oft en complex\, require a lot of data\, and are not always interpretable. An alternative and general approach\, inpired by content-based image retrieva l (Kato\, 1992)\, is to use spatial signatures to represent spatial patter ns and dissimilarity measures to compare them (Jasiewicz and Stepinski\, 2 013).\n\nA spatial signature is any numerical representation (compression) of a spatial pattern. For a categorical raster\, it can be a co-occurrenc e vector of classes in a local window\, while for a time series\, it may b e a vector of values in a given cell. Then\, having spatial signatures for both areas (sensors\, moments)\, we can compare them using a dissimilarit y measure (e.g.\, Euclidean distance\, cosine similarity\, etc.) (Cha\, 20 07). This approach can compare complex\, multidimensional spatial patterns \, but at the same time\, it gives some degree of interpretability. It can also be further applied to many techniques of spatial data analysis\, inc luding spatial clustering (to find groups of areas with similar spatial pa tterns) and segmentation (to create regions with similar spatial patterns) .\n\nWhile the concept of applying spatial signature and dissimilarity mea sures is powerful\, there are still many unresolved issues and questions t o consider. It includes the topics of scale of comparison\, input data res olution\, dimensions\, or types\, used spatial signatures\, and selected d issimilarity metrics. There is still a lack of studies that systematically compare different methods of assessing similarity between spatial pattern s\, or suggest good practices in their use. At the same time\, a growing n umber of FOSS tools allows us to test various methods and apply them to re al-life scenarios.\n\nThe goal of this work is to provide an overview of e xisting R packages for comparing spatial patterns. These include ‘motif ’ (for comparing spatial signatures for categorical rasters\; Nowosad\, 2021)\, ‘spquery’ (allowing for comparing spatial signatures for conti nuous rasters)\, and ‘supercells’ (for segmentation of various types o f spatial rasters based on their patterns\; Nowosad and Stepinski\, 2022). It will show how they can be applied in real-life cases and what their li mitations are. This work also aims to open a discussion about the methods for assessing similarity between spatial patterns and their FOSS implement ations.\n\nReferences\n\nCha\, S-H. (2007). Comprehensive Survey on Distan ce/Similarity Measures Between Probability Density Functions. Int. J. Math . Model. Meth. Appl. Sci.\n\nGustafson\, E.J. (1998) Quantifying landscape spatial pattern: what is the state of the art? Ecosystems\n\nJasiewicz\, J.\, & Stepinski\, T. F. (2013). Example-Based Retrieval of Alike Land-Cov er Scenes From NLCD2006 Database. IEEE Geoscience and Remote Sensing Lette rs\, https://doi.org/10.1109/lgrs.2012.2196019\n\nKato\, T. (1992) Databas e architecture for content-based image retrieval\, Image Storage and Retri eval Systems\, https://doi.org/10.1117/12.58497\n\nNowosad\, J. (2021). Mo tif: an open-source R tool for pattern-based spatial analysis. Landscape E cology\, https://doi.org/10.1007/s10980-020-01135-0\n\nNowosad\, J.\, & St epinski\, T. F. (2022). Extended SLIC superpixels algorithm for applicatio ns to non-imagery geospatial rasters. International Journal of Applied Ear th Observation and Geoinformation\, https://doi.org/10.1016/j.jag.2022.102 935 DTSTAMP:20260525T175603Z LOCATION:Omicum SUMMARY:Comparing spatial patterns in raster data using R - Jakub Nowosad URL:https://talks.staging.osgeo.org/foss4g-europe-2024-academic-track/talk/ DQCQMS/ END:VEVENT END:VCALENDAR