image_correction() function to perform color correction of an image using a set of known color references (e.g., from a color checker)pick_rgb_area() to pick RGB mean values of an area drawn by picking points in the imageobject_scatter() function to plot object thumbnails at x-y coordinates derived from image featuresInclude morphologic operation when foreground and background arguments are used in analyze_objects().
mosaic_classify() function to classify orthomosaics based on break values defined by users.new image_contour_line() function allows users to detect smooth contour lines in an image.image_canny_edge() function, which implements Canny Edge detection.image_line_segment() function that enables automatic line segment detection using the Line Segment Detector (LSD) algorithm.shapefile_build()lwgeom package to compute perimeter.mosaic_crop() now have a type argument, that allows returning a cropped or masked raster.image_autocrop() now has threshold and invert arguments.image_label() to label binary images with options to control a max_gap, ie., the maximum allowable gap (in pixels) between connected components to be considered as part of the same object.mosaic_*() family of functions for high-throughput phenotyping, enabling efficient analysis of large-scale image data.shapefile_*() functions for handling and manipulating shapefiles, providing streamlined spatial data management.mosaic_analyze() to analyze orthomosaics.A new mosaic_*() family of functions to handle orthomosaics from RGB and multispectral images.
object_export() and object_export_shp() to export objects from single images to multiple images.
image_augment() to augment an image by rotating it multiple times.
save_image = TRUE in measure_disease_byl() now exports the processed images to a local directory, allowing the use of sad() call to the created object.analyze_objects(), analyze_objects_shp(), and image_view(), allowing to correctly choose the image band.image_segment_mask() now has a col_background argument.width_at included in analyze_objects().object_export_shp() and object_export() to export objects from an image to multiple images in the current working directory.plot_index_shp() to plot rectangles on top of an RGB image, where each rectangle is colored based on a quantitative variable.poly_center() by calling the column position instead column name (X1)analyze_objects_iter() to execute an interactive section of analyze_objects().
measure_disease_byl() to measure disease severity 'by leaf' in an image with several leaves.
object_split() to split multiples objects of an image into a list of images.
pca(), plot.pca(), get_biplot() as helper functions to perform Principal Component Analysis.
rownames_to_column(), column_to_rownames(), separate_col(), round_cols() as helper functions to manipulate data.
A set of poly_*() function to analyze polygons. All of them are based on a set of coordinate points describing the edge of the object(s). See ?utils_polygon for more details.
get_wd_here() and set_wd_here() to deal with working directories.
apply_fun_to_imgs() to apply a function (or functions) to a set of images stored in the working directory.
make_brush(), make_mask(), and image_segment_mask() to create masks and segment images based on such a mask.
image_segment_manual(), image segment kmeans(), and image_segment_mask() to perform image segmentation in different ways.
A new family of efourier_*() functions to performs Elliptical Fourier Analysis.
efourier(): Elliptical Fourier Analysisefourier_coefs(): Get Fourier coefficientsefourier_error(): Erros between the original and reconstructed outlineefourier_inv(): Inverse Elliptical Fourier Analysisefourier_norm(): Normalized Fourier coefficientsefourier_power(): Power in Fourier Analysisefourier_shape(): Draw shapes based on Fourier coefficientsA new family of landmarks_*() functions to handle landmarks
landmarks(): Create image landmarkslandmarks_add(): Artificially inflates the number of landmarkslandmarks_angle(): Angles between landmarkslandmarks_dist(): Distances between landmarkslandmarks_regradi(): Pseudolandmarks with equally spaced anglesobject_edge() to detect edges in images using Sobel-Feldman Operator.
A new family of *_shp() functions to analyze shape files.
image_shp() to construct a shape file from an image.object_split_shp() to splits image objects based on a shapefile.analyze_objects_shp() to analyze objects using shapefiles.measure_disease_shp() to measure disease using shapefiles.New plot_index() function to plot an image index using raster package, and optionaly using the mapview package to show the image index.
New image_view() function to create an interactive map view of an image. This function allows users to interactively edit and analyze an image using mapview and mapedit packages.
New image_prepare_mv() function to prepare an image to be analyzed for analyze_objects_shp(). This function aligns and crops the image using either base or mapview visualization.
New viewer option added. Now, iterative functions such as pick_palette() and measure_disease_iter() have an argument viewer. If not provided, the value is retrieved using get_pliman_viewer(). This option controls the type of viewer to use for interactive plotting. The available options are "base" and "mapview". If set to "base", the base R graphics system is used for interactive plotting. If set to "mapview", the mapview package is used, allowing the users to draw shapes like points and polygons with mapedit package. To set this argument globally for all functions in the package, you can use the set_pliman_viewer() function. For example, you can run set_pliman_viewer("mapview") to set the viewer option to "mapview" for all functions.
Haralick's features that quantify pixel texture for image objects were included.
Several measures were added in analyze_objects(). The function now wraps some poly_*() functions to compute shape measures such as width, length, elongation, circularity. Haralick's features are now computed by default. . This improvement was at cost of a slight increase in computation time.
analyze_objects(), measure_disease(), and measure_disease_byl() have now a filter argument that applies a median filtering in the binary mask. This is useful to reduce the noise in the segmentation of objects.
Arguments reference_larger and reference_smaller were included in analyze_objects() indicating when the larger/smaller object in the image must be used as the reference object.
Arguments efourier and nharm included in analyze_objects(). If efourier = TRUE, Elliptical Fourier analysis is computed for each object depending on the number of harmonics (nharm).
Logical arguments reference_larger and reference_smaller included in analyze_objects(). Those indicates when the larger/smaller object in the image must be used as the reference object. This only is valid when reference = TRUE and reference_area indicates the area of the reference object. IMPORTANT. When reference_smaller is used, objects with an area smaller than 1% of the mean of all the objects are ignored. This is used to remove possible noise in the image such as dust. So, be sure the reference object has an area that will be not removed by that cutpoint.
Rcpp and RcppArmadillo dependencies were included, allowing the implementation of C++ code. This will dramatically reduce the time computing of some functions/procesures. As an example, we wave.
object_rgb() function to extract the RGB values from an image (1445 x 1084) with ~1400 objects.*_poly() functions.get_measures() now remove known objects from the results when using the id argument.measure in get_measures() now accepts a numeric object stored in the global environment.analyze_objects() now returns the objects object_index and object_rgb when the argument pattern is used. Thanks to João Paulo Oliveira Ribeiro for alerting me regarding this issue.reference in analyze_objects() to adjust measures using a reference object in the image.object_index in analyze_objects() now recognizes the names of built-in indexes (see ?pliman_indexes()).plot.image_index() not limits the number of pixels to reduce plotting time.show_image argument changed with plot to standardize the argument across functions.rgb_to_hsb() optimized using C++.rows and cols with nrow and ncol, respectively, in functions analyze_objects_shp(), image_shp(), measure_disease_shp(), and object_split_shp(), to standardize the arguments across functions.measure_disease_iter() to measure disease in an interactive section.pick_count() to count objects in an image manually.pick_palette() to create an image palette by picking up color point(s) from the imagepick_rgb() to pick up the RGB values from selected point(s) in the image.summary_index() to summary the index either between and within objects.pliman now exports the foward-pipe operator %>%. Code from poorman package.count_objects(), image_show(), leaf_area(), objects_rgb(), prop_segmented(), and symptomatic_area()) were removed.analyze_objects() is now used as the main function to compute the number and shape of objects.measure_disease() is now used as the main function to perform phytopatometry studies. The function can compute symptomatic area, as well as the number and shape of lesions.image_segment_iter() is used to performs iterative image segmentation.conv_hull(), poly_area(), poly_mass(), poly_spline(), plot_contour(), and plot_ellipse() as utilities for analyzing polygons.dpi() to compute the resolution (dots per inch) of an image.tune_tolerance() for tunning the tolerance parameter.objects_rgb() will be depracated in the future. Now, to compute an index for each object use the object_index argument in analyze_objects(), for example, analyze_objects(object_index = "B").leaf_area() will be depracated in the future. Now, combine analyze_objects() with get_measures() to obtain the area and shape of objects (leaves).prop_segmented() is now deprecated in favour of image_segment_iter().count_lesions() is now deprecated. Now, to compute the number and shape of lesions, use the argument show_features = TRUE in measure_disease().image_show() is now deprecated in favour of plot().fill_hull argument in symptomatic_area() and count_lesions()image_contrast() function to avoid error regarding image resolution.subfolder in image_export() to export an image to a subfolder.EBImage installation is checked when pliman is installed.image_pliman() now returns the image object instead of the path to the image. So, it is not necessarily to call it within image_import().image_autocrop() for automatic image cropping.image_filter() to perform median-based filtering.image_contrast() to improve contrast by performing adaptive histogram equalizationobject_coord() to get the object coordinates and (optionally) draw a bounding rectangle around multiple objects in an image.object_id() to get the object identification in an image.object_isolate() to isolate an object from an image.prop_segmented() to perform (iterative) image segmentation with pixels proportion.filter in count_objects() and prop_segmented().pliman package submitted to CRAN.