Abstract
In this post, I am going to analyze the plant leaf dataset as made available by UCI Machine Learning repository at link:
https://archive.ics.uci.edu/ml/datasets/One-hundred+plant+species+leaves+data+set#
Exploratory analysis and models for plant leaf classification will be introduced.
Introduction
The dataset comprises sixteen samples each of one-hundred plant species. Its analysis was introduced within ref. [1]. That paper describes a method designed to work in conditions of small training set size and possibly incomplete extraction of features. This motivated a separate processing of three feature types:
- shape
- texture
- margin
Those are then combined to provide an overall indication of the species (and associated probability). For an accurate description of those feature, please see ref. [1] where the classification is implemented by a K-Nearest-Neighbor density estimator. The authors show the accuracy reached by K-Nearest-Neighbor classification for any combination of the datasets in use (see ref. [1] table 2). The top accuracy is 96% and it is achieved when using all three datasets.
In this post, I introduce an exploratory analysis for those datasets. On next post classification models will be outlined.
Packages
suppressPackageStartupMessages(library(caret))
suppressPackageStartupMessages(library(dplyr))
suppressPackageStartupMessages(library(ggplot2))
suppressPackageStartupMessages(library(corrplot))Exploratory Analysis
First, we download the zip file providing all plant leaf datasets.
url <- "https://archive.ics.uci.edu/ml/machine-learning-databases/00241/100%20leaves%20plant%20species.zip"
temp_file <- tempfile()
download.file(url, temp_file)Actually, the zip file provides further content, such as a readme file, a pdf document and a folder tree providing with a set of images for each plant specie. However, the files we are specifically interested in are:
margin_file <- "100 leaves plant species/data_Mar_64.txt"
shape_file <- "100 leaves plant species/data_Sha_64.txt"
texture_file <- "100 leaves plant species/data_Tex_64.txt"Unzipping the downloaded file into a temporary directory.
files_to_unzip <- c(margin_file, shape_file, texture_file)
unzip(temp_file, files = files_to_unzip, exdir=".", overwrite = TRUE)Reading datasets of interest.
margin_data <- read.csv(margin_file, header=FALSE, sep=",", stringsAsFactors = TRUE)
head(margin_data)## V1 V2 V3 V4 V5 V6 V7
## 1 Acer Campestre 0.003906 0.003906 0.027344 0.033203 0.007812 0.017578
## 2 Acer Campestre 0.005859 0.013672 0.027344 0.025391 0.013672 0.029297
## 3 Acer Campestre 0.011719 0.001953 0.027344 0.044922 0.017578 0.042969
## 4 Acer Campestre 0.013672 0.011719 0.037109 0.017578 0.011719 0.087891
## 5 Acer Campestre 0.007812 0.009766 0.027344 0.025391 0.001953 0.005859
## 6 Acer Campestre 0.015625 0.003906 0.015625 0.046875 0.013672 0.064453
## V8 V9 V10 V11 V12 V13 V14 V15
## 1 0.023438 0.005859 0.000000 0.015625 0.015625 0.015625 0.025391 0.000000
## 2 0.019531 0.000000 0.001953 0.021484 0.007812 0.003906 0.013672 0.003906
## 3 0.023438 0.000000 0.003906 0.019531 0.017578 0.005859 0.009766 0.005859
## 4 0.023438 0.000000 0.000000 0.027344 0.021484 0.009766 0.013672 0.001953
## 5 0.015625 0.000000 0.005859 0.017578 0.039062 0.007812 0.042969 0.001953
## 6 0.017578 0.003906 0.000000 0.033203 0.013672 0.009766 0.023438 0.000000
## V16 V17 V18 V19 V20 V21 V22 V23 V24
## 1 0.015625 0 0.025391 0.027344 0.033203 0.009766 0.025391 0.001953 0
## 2 0.009766 0 0.044922 0.011719 0.007812 0.031250 0.017578 0.001953 0
## 3 0.015625 0 0.027344 0.017578 0.019531 0.019531 0.023438 0.001953 0
## 4 0.013672 0 0.021484 0.017578 0.011719 0.027344 0.009766 0.003906 0
## 5 0.007812 0 0.017578 0.027344 0.025391 0.003906 0.019531 0.003906 0
## 6 0.013672 0 0.031250 0.027344 0.017578 0.015625 0.011719 0.000000 0
## V25 V26 V27 V28 V29 V30 V31 V32
## 1 0.003906 0.001953 0.044922 0.000000 0.031250 0.033203 0.015625 0.003906
## 2 0.001953 0.001953 0.037109 0.000000 0.029297 0.025391 0.039062 0.003906
## 3 0.011719 0.001953 0.042969 0.000000 0.011719 0.023438 0.035156 0.001953
## 4 0.011719 0.000000 0.050781 0.001953 0.019531 0.013672 0.011719 0.003906
## 5 0.009766 0.003906 0.017578 0.001953 0.009766 0.031250 0.015625 0.001953
## 6 0.007812 0.000000 0.027344 0.000000 0.015625 0.027344 0.031250 0.001953
## V33 V34 V35 V36 V37 V38 V39 V40
## 1 0.000000 0.013672 0.001953 0.015625 0.015625 0.017578 0.033203 0.015625
## 2 0.000000 0.035156 0.000000 0.027344 0.015625 0.019531 0.029297 0.023438
## 3 0.001953 0.027344 0.000000 0.003906 0.017578 0.027344 0.033203 0.007812
## 4 0.000000 0.027344 0.000000 0.013672 0.007812 0.017578 0.035156 0.021484
## 5 0.001953 0.037109 0.000000 0.011719 0.037109 0.015625 0.039062 0.009766
## 6 0.001953 0.013672 0.003906 0.015625 0.027344 0.001953 0.021484 0.015625
## V41 V42 V43 V44 V45 V46 V47 V48
## 1 0.003906 0.003906 0.031250 0.035156 0.015625 0.044922 0.007812 0.029297
## 2 0.005859 0.003906 0.017578 0.025391 0.011719 0.025391 0.005859 0.035156
## 3 0.001953 0.007812 0.035156 0.015625 0.011719 0.027344 0.003906 0.011719
## 4 0.000000 0.003906 0.035156 0.027344 0.015625 0.029297 0.007812 0.011719
## 5 0.001953 0.000000 0.050781 0.023438 0.005859 0.058594 0.003906 0.015625
## 6 0.003906 0.001953 0.025391 0.029297 0.013672 0.039062 0.009766 0.025391
## V49 V50 V51 V52 V53 V54 V55 V56
## 1 0.037109 0.000000 0.027344 0.005859 0.001953 0.041016 0.000000 0.011719
## 2 0.033203 0.000000 0.015625 0.011719 0.003906 0.037109 0.001953 0.017578
## 3 0.019531 0.001953 0.013672 0.005859 0.007812 0.068359 0.003906 0.035156
## 4 0.023438 0.000000 0.017578 0.009766 0.007812 0.052734 0.003906 0.015625
## 5 0.021484 0.001953 0.015625 0.023438 0.005859 0.027344 0.000000 0.023438
## 6 0.031250 0.000000 0.019531 0.001953 0.000000 0.050781 0.003906 0.005859
## V57 V58 V59 V60 V61 V62 V63 V64
## 1 0.000000 0.005859 0.035156 0.027344 0.033203 0.001953 0.000000 0.017578
## 2 0.000000 0.021484 0.017578 0.046875 0.005859 0.003906 0.003906 0.046875
## 3 0.000000 0.015625 0.021484 0.056641 0.009766 0.003906 0.000000 0.015625
## 4 0.001953 0.021484 0.029297 0.033203 0.003906 0.000000 0.001953 0.027344
## 5 0.001953 0.021484 0.048828 0.056641 0.019531 0.000000 0.000000 0.013672
## 6 0.001953 0.021484 0.017578 0.041016 0.017578 0.003906 0.001953 0.041016
## V65
## 1 0
## 2 0
## 3 0
## 4 0
## 5 0
## 6 0shape_data <- read.csv(shape_file, header=FALSE, sep=",", stringsAsFactors = TRUE)
head(shape_data)## V1 V2 V3 V4 V5 V6
## 1 Acer Capillipes 0.00057884 0.00060866 0.00055063 0.00055412 0.00060251
## 2 Acer Capillipes 0.00063028 0.00066074 0.00071871 0.00065149 0.00064287
## 3 Acer Capillipes 0.00061634 0.00061527 0.00060560 0.00056823 0.00055835
## 4 Acer Capillipes 0.00061271 0.00056928 0.00056431 0.00060722 0.00064275
## 5 Acer Capillipes 0.00059946 0.00055240 0.00055795 0.00056888 0.00061560
## 6 Acer Capillipes 0.00055321 0.00058276 0.00052074 0.00057495 0.00060107
## V7 V8 V9 V10 V11 V12
## 1 0.00061420 0.00061099 0.00061126 0.00061127 0.00059401 0.00056723
## 2 0.00063999 0.00064591 0.00062378 0.00058375 0.00054570 0.00052020
## 3 0.00055227 0.00055119 0.00055162 0.00053110 0.00052965 0.00050656
## 4 0.00064712 0.00066345 0.00065759 0.00063546 0.00060048 0.00058617
## 5 0.00063898 0.00063149 0.00063393 0.00063895 0.00059634 0.00058721
## 6 0.00060068 0.00060566 0.00059552 0.00059341 0.00056913 0.00054535
## V13 V14 V15 V16 V17 V18
## 1 0.00054544 0.00054852 0.00052220 0.00051996 0.00050029 0.00049550
## 2 0.00049217 0.00047365 0.00045806 0.00044512 0.00041131 0.00041642
## 3 0.00050098 0.00048780 0.00042640 0.00044484 0.00047091 0.00045628
## 4 0.00056935 0.00054666 0.00051782 0.00050056 0.00048341 0.00046331
## 5 0.00057578 0.00056669 0.00053864 0.00052719 0.00051216 0.00050576
## 6 0.00054417 0.00052943 0.00052480 0.00050725 0.00049601 0.00048730
## V19 V20 V21 V22 V23 V24
## 1 0.00048506 0.00043270 0.00043344 0.00044133 0.00045759 0.00047369
## 2 0.00041792 0.00043338 0.00046667 0.00049444 0.00051226 0.00051463
## 3 0.00047391 0.00050035 0.00053424 0.00053726 0.00058021 0.00059195
## 4 0.00044204 0.00043805 0.00044337 0.00047125 0.00048651 0.00051293
## 5 0.00047713 0.00042037 0.00043828 0.00044307 0.00046131 0.00048500
## 6 0.00043755 0.00043060 0.00043600 0.00044354 0.00047238 0.00048704
## V25 V26 V27 V28 V29 V30
## 1 0.00050027 0.00051038 0.00055370 0.00057409 0.00061545 0.00063420
## 2 0.00054706 0.00056805 0.00061162 0.00062314 0.00067340 0.00070644
## 3 0.00063824 0.00065655 0.00069176 0.00075707 0.00081739 0.00087267
## 4 0.00050608 0.00054728 0.00057562 0.00061750 0.00060790 0.00066313
## 5 0.00050022 0.00050812 0.00054919 0.00057575 0.00062919 0.00065269
## 6 0.00050947 0.00052215 0.00055134 0.00057878 0.00061888 0.00067277
## V31 V32 V33 V34 V35 V36
## 1 0.00066996 0.00070086 0.00076520 0.00082210 0.00088670 0.00095273
## 2 0.00074835 0.00079712 0.00085795 0.00091375 0.00097812 0.00095934
## 3 0.00093889 0.00091932 0.00084868 0.00079460 0.00073694 0.00067678
## 4 0.00069860 0.00074795 0.00080933 0.00087363 0.00094199 0.00097837
## 5 0.00069252 0.00072327 0.00078248 0.00083414 0.00088673 0.00095142
## 6 0.00072001 0.00075503 0.00080892 0.00086298 0.00091485 0.00097837
## V37 V38 V39 V40 V41 V42
## 1 0.00089004 0.00082684 0.00077746 0.00072624 0.00068827 0.00064738
## 2 0.00089228 0.00082245 0.00076643 0.00069780 0.00066379 0.00063080
## 3 0.00064251 0.00060182 0.00058021 0.00056083 0.00051736 0.00050721
## 4 0.00091476 0.00085575 0.00079947 0.00074441 0.00070522 0.00066982
## 5 0.00095801 0.00088576 0.00081584 0.00074559 0.00069396 0.00064167
## 6 0.00091097 0.00084917 0.00078448 0.00072748 0.00069029 0.00064487
## V43 V44 V45 V46 V47 V48
## 1 0.00060111 0.00059136 0.00054355 0.00051874 0.00051434 0.00048473
## 2 0.00058542 0.00055536 0.00050568 0.00048712 0.00045827 0.00044403
## 3 0.00048000 0.00044296 0.00044281 0.00044033 0.00047195 0.00048522
## 4 0.00063146 0.00059858 0.00056835 0.00052424 0.00051716 0.00048934
## 5 0.00063091 0.00059208 0.00056242 0.00051641 0.00053070 0.00050384
## 6 0.00059307 0.00056657 0.00052382 0.00050103 0.00047318 0.00047589
## V49 V50 V51 V52 V53 V54
## 1 0.00048040 0.00045491 0.00045303 0.00044640 0.00042522 0.00047467
## 2 0.00041724 0.00041694 0.00041162 0.00041296 0.00045280 0.00046123
## 3 0.00050598 0.00052326 0.00054469 0.00054883 0.00055364 0.00056865
## 4 0.00047944 0.00046101 0.00044595 0.00045062 0.00049919 0.00049343
## 5 0.00047250 0.00045785 0.00044651 0.00043924 0.00049357 0.00049833
## 6 0.00045528 0.00045104 0.00044349 0.00043897 0.00044078 0.00049003
## V55 V56 V57 V58 V59 V60
## 1 0.00048917 0.00050727 0.00053275 0.00055509 0.00056515 0.00058112
## 2 0.00047762 0.00050331 0.00051962 0.00053258 0.00056427 0.00059566
## 3 0.00059373 0.00058093 0.00058984 0.00058882 0.00056550 0.00057454
## 4 0.00050075 0.00051610 0.00053550 0.00054904 0.00054225 0.00056617
## 5 0.00050411 0.00052932 0.00054267 0.00055699 0.00056183 0.00055837
## 6 0.00050851 0.00052437 0.00052753 0.00054212 0.00057232 0.00056827
## V61 V62 V63 V64 V65
## 1 0.00059677 0.00062541 0.00062410 0.00061671 0.00061401
## 2 0.00062344 0.00064182 0.00066119 0.00067058 0.00066691
## 3 0.00061797 0.00054334 0.00059248 0.00060658 0.00060235
## 4 0.00059244 0.00060052 0.00060897 0.00061372 0.00060271
## 5 0.00059111 0.00060751 0.00061317 0.00061027 0.00059428
## 6 0.00061607 0.00062150 0.00062151 0.00062510 0.00060623texture_data <- read.csv(texture_file, header=FALSE, sep=",", stringsAsFactors = TRUE)
head(texture_data)## V1 V2 V3 V4 V5 V6 V7 V8
## 1 Acer Campestre 0.025391 0.012695 0.003906 0.004883 0.039062 0 0.017578
## 2 Acer Campestre 0.004883 0.018555 0.002930 0.000000 0.069336 0 0.013672
## 3 Acer Campestre 0.018555 0.013672 0.002930 0.002930 0.051758 0 0.019531
## 4 Acer Campestre 0.035156 0.023438 0.000977 0.000000 0.061523 0 0.021484
## 5 Acer Campestre 0.038086 0.014648 0.003906 0.000977 0.046875 0 0.022461
## 6 Acer Campestre 0.045898 0.017578 0.000000 0.003906 0.044922 0 0.017578
## V9 V10 V11 V12 V13 V14 V15 V16 V17
## 1 0.035156 0.023438 0.013672 0.000977 0.036133 0 0.000000 0 0
## 2 0.043945 0.026367 0.000000 0.000000 0.395510 0 0.000000 0 0
## 3 0.035156 0.022461 0.000977 0.000000 0.208980 0 0.000000 0 0
## 4 0.061523 0.010742 0.001953 0.000000 0.222660 0 0.000000 0 0
## 5 0.053711 0.019531 0.004883 0.000977 0.076172 0 0.000000 0 0
## 6 0.036133 0.006836 0.002930 0.000000 0.058594 0 0.000977 0 0
## V18 V19 V20 V21 V22 V23 V24 V25 V26
## 1 0.024414 0 0.028320 0.021484 0 0.003906 0.007812 0.001953 0.00000
## 2 0.032227 0 0.010742 0.027344 0 0.000000 0.032227 0.000977 0.00000
## 3 0.042969 0 0.017578 0.021484 0 0.000000 0.035156 0.003906 0.00000
## 4 0.022461 0 0.008789 0.018555 0 0.000000 0.036133 0.000000 0.00000
## 5 0.019531 0 0.006836 0.035156 0 0.007812 0.033203 0.000000 0.00000
## 6 0.021484 0 0.011719 0.024414 0 0.000000 0.036133 0.003906 0.00293
## V27 V28 V29 V30 V31 V32 V33 V34 V35
## 1 0.001953 0.044922 0 0.027344 0.000000 0.031250 0 0 0.069336
## 2 0.000000 0.000977 0 0.028320 0.000977 0.027344 0 0 0.018555
## 3 0.000000 0.014648 0 0.023438 0.000000 0.025391 0 0 0.062500
## 4 0.000000 0.000977 0 0.035156 0.000000 0.040039 0 0 0.078125
## 5 0.000000 0.015625 0 0.032227 0.000000 0.049805 0 0 0.073242
## 6 0.000000 0.042969 0 0.018555 0.000000 0.064453 0 0 0.100590
## V36 V37 V38 V39 V40 V41 V42 V43 V44
## 1 0.000000 0 0.000977 0.027344 0.023438 0.012695 0.006836 0 0.011719
## 2 0.003906 0 0.000000 0.020508 0.000000 0.000977 0.000000 0 0.000977
## 3 0.002930 0 0.000000 0.027344 0.004883 0.007812 0.000000 0 0.012695
## 4 0.002930 0 0.000000 0.039062 0.001953 0.004883 0.000000 0 0.000977
## 5 0.000000 0 0.000000 0.019531 0.016602 0.006836 0.000000 0 0.012695
## 6 0.000977 0 0.000000 0.025391 0.011719 0.008789 0.002930 0 0.023438
## V45 V46 V47 V48 V49 V50 V51 V52 V53 V54
## 1 0.162110 0.005859 0.022461 0.025391 0.013672 0 0.041016 0 0 0
## 2 0.003906 0.031250 0.014648 0.012695 0.037109 0 0.011719 0 0 0
## 3 0.020508 0.033203 0.016602 0.025391 0.021484 0 0.007812 0 0 0
## 4 0.004883 0.019531 0.025391 0.021484 0.019531 0 0.012695 0 0 0
## 5 0.058594 0.018555 0.029297 0.020508 0.013672 0 0.035156 0 0 0
## 6 0.082031 0.021484 0.031250 0.032227 0.016602 0 0.004883 0 0 0
## V55 V56 V57 V58 V59 V60 V61 V62 V63 V64
## 1 0.012695 0.103520 0 0.001953 0.000977 0.022461 0 0 0.001953 0
## 2 0.011719 0.070312 0 0.017578 0.000000 0.004883 0 0 0.000000 0
## 3 0.022461 0.156250 0 0.008789 0.000000 0.001953 0 0 0.000000 0
## 4 0.009766 0.105470 0 0.026367 0.000000 0.002930 0 0 0.000000 0
## 5 0.020508 0.150390 0 0.002930 0.000000 0.023438 0 0 0.000000 0
## 6 0.018555 0.110350 0 0.006836 0.000000 0.008789 0 0 0.000977 0
## V65
## 1 0.027344
## 2 0.002930
## 3 0.005859
## 4 0.022461
## 5 0.015625
## 6 0.030273We should expect to find equal dimensions for all three datasets.
dim(margin_data)## [1] 1600 65dim(shape_data)## [1] 1600 65dim(texture_data)## [1] 1599 65However, the texture dataset has one row less than other ones. We will fix such issue at a later moment.
We then label the columns for each dataset in order to have proper headers.
n_features <- ncol(margin_data) - 1
colnames(margin_data) <- c("species", paste("margin", as.character(1:n_features), sep=""))
margin_data$species <- factor(margin_data$species)
sum(complete.cases(margin_data)) == nrow(margin_data)## [1] TRUEdim(margin_data)## [1] 1600 65length(unique(margin_data$species))## [1] 100n_features <- ncol(shape_data) - 1
colnames(shape_data) <- c("species", paste("shape", as.character(1:n_features), sep=""))
shape_data$species <- factor(shape_data$species)
sum(complete.cases(shape_data)) == nrow(shape_data)## [1] TRUEdim(shape_data)## [1] 1600 65length(unique(shape_data$species))## [1] 100n_features <- ncol(texture_data) - 1
colnames(texture_data) <- c("species", paste("texture", as.character(1:n_features), sep=""))
texture_data$species <- factor(texture_data$species)
sum(complete.cases(texture_data)) == nrow(texture_data)## [1] TRUEdim(texture_data)## [1] 1599 65length(unique(texture_data$species))## [1] 100To understand which row is missing in the texture dataset, we split each of those as based on species.
margin_count <- sapply(base::split(margin_data, margin_data$species), nrow)
shape_count <- sapply(base::split(shape_data, shape_data$species), nrow)
texture_count <- sapply(base::split(texture_data, texture_data$species), nrow)which(margin_count != texture_count)## Acer Campestre
## 1which(shape_count != texture_count)## Acer Campestre
## 1There is a missing entry for Acer Campestre in texture dataset, issue that we will fix at a later moment.
Summaries can be run in order to have an idea of margin, shape and texture range values across all the species.
summary(margin_data)## species margin1 margin2
## Acer Campestre : 16 Min. :0.000000 Min. :0.000000
## Acer Capillipes: 16 1st Qu.:0.001953 1st Qu.:0.001953
## Acer Circinatum: 16 Median :0.009766 Median :0.011719
## Acer Mono : 16 Mean :0.017423 Mean :0.028365
## Acer Opalus : 16 3rd Qu.:0.025391 3rd Qu.:0.039551
## Acer Palmatum : 16 Max. :0.087891 Max. :0.205080
## (Other) :1504
## margin3 margin4 margin5 margin6
## Min. :0.00000 Min. :0.000000 Min. :0.000000 Min. :0.00000
## 1st Qu.:0.01367 1st Qu.:0.005859 1st Qu.:0.001953 1st Qu.:0.00000
## Median :0.02344 Median :0.013672 Median :0.007812 Median :0.01367
## Mean :0.03189 Mean :0.023022 Mean :0.014315 Mean :0.03808
## 3rd Qu.:0.04297 3rd Qu.:0.029297 3rd Qu.:0.019531 3rd Qu.:0.05664
## Max. :0.16797 Max. :0.169920 Max. :0.111330 Max. :0.31055
##
## margin7 margin8 margin9
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.005859 1st Qu.:0.000000 1st Qu.:0.001953
## Median :0.015625 Median :0.000000 Median :0.005859
## Mean :0.019226 Mean :0.001084 Mean :0.007098
## 3rd Qu.:0.029297 3rd Qu.:0.000000 3rd Qu.:0.007812
## Max. :0.091797 Max. :0.031250 Max. :0.083984
##
## margin10 margin11 margin12
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.005859 1st Qu.:0.003906 1st Qu.:0.001953
## Median :0.015625 Median :0.013672 Median :0.007812
## Mean :0.018739 Mean :0.024265 Mean :0.012054
## 3rd Qu.:0.027344 3rd Qu.:0.041016 3rd Qu.:0.019531
## Max. :0.097656 Max. :0.125000 Max. :0.056641
##
## margin13 margin14 margin15
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.009766 1st Qu.:0.000000 1st Qu.:0.001953
## Median :0.027344 Median :0.001953 Median :0.011719
## Mean :0.041259 Mean :0.007704 Mean :0.015629
## 3rd Qu.:0.062500 3rd Qu.:0.007812 3rd Qu.:0.025391
## Max. :0.416020 Max. :0.082031 Max. :0.066406
##
## margin16 margin17 margin18
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.005859 1st Qu.:0.005859
## Median :0.000000 Median :0.013672 Median :0.015625
## Mean :0.000155 Mean :0.015164 Mean :0.020378
## 3rd Qu.:0.000000 3rd Qu.:0.021484 3rd Qu.:0.029297
## Max. :0.029297 Max. :0.064453 Max. :0.359380
##
## margin19 margin20 margin21
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.003906 1st Qu.:0.005859 1st Qu.:0.003906
## Median :0.007812 Median :0.011719 Median :0.013672
## Mean :0.012681 Mean :0.013242 Mean :0.019042
## 3rd Qu.:0.017578 3rd Qu.:0.017578 3rd Qu.:0.031250
## Max. :0.125000 Max. :0.056641 Max. :0.101560
##
## margin22 margin23 margin24
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.001953 Median :0.000000 Median :0.001953
## Mean :0.006163 Mean :0.001034 Mean :0.007511
## 3rd Qu.:0.007812 3rd Qu.:0.000000 3rd Qu.:0.009766
## Max. :0.062500 Max. :0.046875 Max. :0.078125
##
## margin25 margin26 margin27
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.009766 1st Qu.:0.000000
## Median :0.001953 Median :0.017578 Median :0.000000
## Mean :0.008899 Mean :0.018656 Mean :0.005560
## 3rd Qu.:0.013672 3rd Qu.:0.025391 3rd Qu.:0.005859
## Max. :0.082031 Max. :0.085938 Max. :0.082031
##
## margin28 margin29 margin30
## Min. :0.000000 Min. :0.00000 Min. :0.000000
## 1st Qu.:0.003906 1st Qu.:0.01172 1st Qu.:0.005859
## Median :0.011719 Median :0.02344 Median :0.011719
## Mean :0.015464 Mean :0.02844 Mean :0.016312
## 3rd Qu.:0.023438 3rd Qu.:0.04102 3rd Qu.:0.021484
## Max. :0.080078 Max. :0.14844 Max. :0.119140
##
## margin31 margin32 margin33
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.005859
## Median :0.003906 Median :0.001953 Median :0.015625
## Mean :0.010898 Mean :0.009730 Mean :0.019579
## 3rd Qu.:0.013672 3rd Qu.:0.009766 3rd Qu.:0.029297
## Max. :0.113280 Max. :0.126950 Max. :0.083984
##
## margin34 margin35 margin36
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.001953 1st Qu.:0.003906
## Median :0.000000 Median :0.009766 Median :0.013672
## Mean :0.001135 Mean :0.013522 Mean :0.017993
## 3rd Qu.:0.001953 3rd Qu.:0.021484 3rd Qu.:0.027344
## Max. :0.029297 Max. :0.082031 Max. :0.089844
##
## margin37 margin38 margin39
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.007812 1st Qu.:0.009766 1st Qu.:0.003906
## Median :0.013672 Median :0.028320 Median :0.011719
## Mean :0.016080 Mean :0.030906 Mean :0.014969
## 3rd Qu.:0.021484 3rd Qu.:0.046875 3rd Qu.:0.023438
## Max. :0.068359 Max. :0.140630 Max. :0.066406
##
## margin40 margin41 margin42
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.005859
## Median :0.003906 Median :0.001953 Median :0.013672
## Mean :0.008188 Mean :0.010566 Mean :0.017839
## 3rd Qu.:0.011719 3rd Qu.:0.009766 3rd Qu.:0.025391
## Max. :0.072266 Max. :0.208980 Max. :0.091797
##
## margin43 margin44 margin45
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.003906 1st Qu.:0.003906 1st Qu.:0.005859
## Median :0.009766 Median :0.009766 Median :0.017578
## Mean :0.018823 Mean :0.012565 Mean :0.024922
## 3rd Qu.:0.027344 3rd Qu.:0.017578 3rd Qu.:0.035156
## Max. :0.117190 Max. :0.064453 Max. :0.177730
##
## margin46 margin47 margin48
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.005859 1st Qu.:0.001953
## Median :0.003906 Median :0.021484 Median :0.017578
## Mean :0.010102 Mean :0.025510 Mean :0.027270
## 3rd Qu.:0.015625 3rd Qu.:0.041016 3rd Qu.:0.044922
## Max. :0.089844 Max. :0.105470 Max. :0.183590
##
## margin49 margin50 margin51
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.005859
## Median :0.001953 Median :0.005859 Median :0.019531
## Mean :0.008529 Mean :0.013793 Mean :0.025535
## 3rd Qu.:0.009766 3rd Qu.:0.021484 3rd Qu.:0.039062
## Max. :0.107420 Max. :0.091797 Max. :0.130860
##
## margin52 margin53 margin54
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.009766 1st Qu.:0.000000
## Median :0.000000 Median :0.021484 Median :0.002929
## Mean :0.002682 Mean :0.024270 Mean :0.010022
## 3rd Qu.:0.001953 3rd Qu.:0.035156 3rd Qu.:0.013672
## Max. :0.083984 Max. :0.097656 Max. :0.128910
##
## margin55 margin56 margin57
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.001953 1st Qu.:0.000000 1st Qu.:0.005859
## Median :0.011719 Median :0.001953 Median :0.009766
## Mean :0.018063 Mean :0.005997 Mean :0.012411
## 3rd Qu.:0.027344 3rd Qu.:0.007812 3rd Qu.:0.017578
## Max. :0.140630 Max. :0.054688 Max. :0.062500
##
## margin58 margin59 margin60
## Min. :0.000000 Min. :0.00000 Min. :0.000000
## 1st Qu.:0.005859 1st Qu.:0.01172 1st Qu.:0.003906
## Median :0.015625 Median :0.02344 Median :0.007812
## Mean :0.020353 Mean :0.03110 Mean :0.011927
## 3rd Qu.:0.029297 3rd Qu.:0.03906 3rd Qu.:0.015625
## Max. :0.113280 Max. :0.25195 Max. :0.089844
##
## margin61 margin62 margin63
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.009766
## Median :0.000000 Median :0.000000 Median :0.021484
## Mean :0.001387 Mean :0.004972 Mean :0.025350
## 3rd Qu.:0.001953 3rd Qu.:0.003906 3rd Qu.:0.037109
## Max. :0.042969 Max. :0.076172 Max. :0.125000
##
## margin64
## Min. :0.000000
## 1st Qu.:0.000000
## Median :0.001953
## Mean :0.004163
## 3rd Qu.:0.003906
## Max. :0.089844
## summary(shape_data)## species shape1 shape2
## Acer Campestre : 16 Min. :0.0001680 Min. :0.0001816
## Acer Capillipes: 16 1st Qu.:0.0005203 1st Qu.:0.0005067
## Acer Circinatum: 16 Median :0.0007098 Median :0.0006896
## Acer Mono : 16 Mean :0.0007366 Mean :0.0007152
## Acer Opalus : 16 3rd Qu.:0.0009275 3rd Qu.:0.0008980
## Acer Palmatum : 16 Max. :0.0023897 Max. :0.0022466
## (Other) :1504
## shape3 shape4 shape5
## Min. :0.0001482 Min. :0.0001037 Min. :0.0001198
## 1st Qu.:0.0004968 1st Qu.:0.0004832 1st Qu.:0.0004782
## Median :0.0006709 Median :0.0006516 Median :0.0006365
## Mean :0.0006902 Mean :0.0006669 Mean :0.0006457
## 3rd Qu.:0.0008640 3rd Qu.:0.0008330 3rd Qu.:0.0007902
## Max. :0.0021123 Max. :0.0019977 Max. :0.0021510
##
## shape6 shape7 shape8
## Min. :0.0001183 Min. :8.493e-05 Min. :6.811e-05
## 1st Qu.:0.0004651 1st Qu.:4.554e-04 1st Qu.:4.409e-04
## Median :0.0006180 Median :5.992e-04 Median :5.833e-04
## Mean :0.0006275 Mean :6.124e-04 Mean :5.998e-04
## 3rd Qu.:0.0007536 3rd Qu.:7.246e-04 3rd Qu.:7.050e-04
## Max. :0.0022666 Max. :2.305e-03 Max. :2.439e-03
##
## shape9 shape10 shape11
## Min. :8.752e-05 Min. :0.0001088 Min. :7.599e-05
## 1st Qu.:4.137e-04 1st Qu.:0.0003952 1st Qu.:3.766e-04
## Median :5.603e-04 Median :0.0005323 Median :5.006e-04
## Mean :5.871e-04 Mean :0.0005751 Mean :5.656e-04
## 3rd Qu.:6.893e-04 3rd Qu.:0.0006816 3rd Qu.:6.691e-04
## Max. :2.521e-03 Max. :0.0026874 Max. :2.774e-03
##
## shape12 shape13 shape14
## Min. :0.0001063 Min. :8.088e-05 Min. :9.678e-05
## 1st Qu.:0.0003582 1st Qu.:3.360e-04 1st Qu.:3.246e-04
## Median :0.0004754 Median :4.552e-04 Median :4.423e-04
## Mean :0.0005580 Mean :5.522e-04 Mean :5.475e-04
## 3rd Qu.:0.0006587 3rd Qu.:6.546e-04 3rd Qu.:6.391e-04
## Max. :0.0028893 Max. :3.007e-03 Max. :2.815e-03
##
## shape15 shape16 shape17
## Min. :7.069e-05 Min. :6.719e-05 Min. :4.453e-05
## 1st Qu.:3.157e-04 1st Qu.:3.104e-04 1st Qu.:3.076e-04
## Median :4.305e-04 Median :4.220e-04 Median :4.255e-04
## Mean :5.429e-04 Mean :5.409e-04 Mean :5.410e-04
## 3rd Qu.:6.363e-04 3rd Qu.:6.350e-04 3rd Qu.:6.349e-04
## Max. :2.719e-03 Max. :2.532e-03 Max. :2.410e-03
##
## shape18 shape19 shape20
## Min. :0.0000376 Min. :2.999e-05 Min. :3.977e-05
## 1st Qu.:0.0003120 1st Qu.:3.196e-04 1st Qu.:3.321e-04
## Median :0.0004272 Median :4.340e-04 Median :4.476e-04
## Mean :0.0005426 Mean :5.439e-04 Mean :5.481e-04
## 3rd Qu.:0.0006301 3rd Qu.:6.244e-04 3rd Qu.:6.208e-04
## Max. :0.0025374 Max. :2.475e-03 Max. :2.383e-03
##
## shape21 shape22 shape23
## Min. :5.501e-05 Min. :4.503e-05 Min. :4.638e-05
## 1st Qu.:3.501e-04 1st Qu.:3.734e-04 1st Qu.:4.022e-04
## Median :4.712e-04 Median :4.933e-04 Median :5.189e-04
## Mean :5.556e-04 Mean :5.650e-04 Mean :5.767e-04
## 3rd Qu.:6.323e-04 3rd Qu.:6.486e-04 3rd Qu.:6.662e-04
## Max. :2.274e-03 Max. :2.420e-03 Max. :2.553e-03
##
## shape24 shape25 shape26
## Min. :5.573e-05 Min. :5.972e-05 Min. :3.117e-05
## 1st Qu.:4.184e-04 1st Qu.:4.355e-04 1st Qu.:4.569e-04
## Median :5.475e-04 Median :5.713e-04 Median :5.957e-04
## Mean :5.882e-04 Mean :6.009e-04 Mean :6.149e-04
## 3rd Qu.:6.786e-04 3rd Qu.:7.093e-04 3rd Qu.:7.338e-04
## Max. :2.434e-03 Max. :2.274e-03 Max. :2.130e-03
##
## shape27 shape28 shape29
## Min. :8.798e-05 Min. :8.671e-05 Min. :5.583e-05
## 1st Qu.:4.687e-04 1st Qu.:4.745e-04 1st Qu.:4.799e-04
## Median :6.162e-04 Median :6.420e-04 Median :6.643e-04
## Mean :6.290e-04 Mean :6.449e-04 Mean :6.632e-04
## 3rd Qu.:7.614e-04 3rd Qu.:7.965e-04 3rd Qu.:8.299e-04
## Max. :1.988e-03 Max. :1.833e-03 Max. :1.824e-03
##
## shape30 shape31 shape32
## Min. :5.366e-05 Min. :6.743e-05 Min. :5.315e-05
## 1st Qu.:4.930e-04 1st Qu.:5.096e-04 1st Qu.:5.214e-04
## Median :6.838e-04 Median :7.018e-04 Median :7.199e-04
## Mean :6.851e-04 Mean :7.083e-04 Mean :7.300e-04
## 3rd Qu.:8.623e-04 3rd Qu.:8.952e-04 3rd Qu.:9.243e-04
## Max. :1.855e-03 Max. :2.002e-03 Max. :2.124e-03
##
## shape33 shape34 shape35
## Min. :9.131e-05 Min. :5.456e-05 Min. :3.387e-05
## 1st Qu.:5.213e-04 1st Qu.:5.121e-04 1st Qu.:5.025e-04
## Median :7.223e-04 Median :7.093e-04 Median :6.760e-04
## Mean :7.345e-04 Mean :7.136e-04 Mean :6.858e-04
## 3rd Qu.:9.313e-04 3rd Qu.:8.991e-04 3rd Qu.:8.628e-04
## Max. :2.111e-03 Max. :1.988e-03 Max. :1.868e-03
##
## shape36 shape37 shape38
## Min. :4.618e-05 Min. :6.663e-05 Min. :6.401e-05
## 1st Qu.:4.850e-04 1st Qu.:4.741e-04 1st Qu.:4.590e-04
## Median :6.544e-04 Median :6.378e-04 Median :6.195e-04
## Mean :6.608e-04 Mean :6.382e-04 Mean :6.174e-04
## 3rd Qu.:8.276e-04 3rd Qu.:7.878e-04 3rd Qu.:7.514e-04
## Max. :1.886e-03 Max. :2.011e-03 Max. :2.107e-03
##
## shape39 shape40 shape41
## Min. :0.0000623 Min. :6.818e-05 Min. :7.033e-05
## 1st Qu.:0.0004462 1st Qu.:4.290e-04 1st Qu.:4.162e-04
## Median :0.0005905 Median :5.677e-04 Median :5.504e-04
## Mean :0.0005989 Mean :5.817e-04 Mean :5.679e-04
## 3rd Qu.:0.0007252 3rd Qu.:7.019e-04 3rd Qu.:6.873e-04
## Max. :0.0021057 Max. :2.195e-03 Max. :2.230e-03
##
## shape42 shape43 shape44
## Min. :5.189e-05 Min. :3.776e-05 Min. :3.532e-05
## 1st Qu.:3.964e-04 1st Qu.:3.744e-04 1st Qu.:3.533e-04
## Median :5.281e-04 Median :5.084e-04 Median :4.894e-04
## Mean :5.579e-04 Mean :5.517e-04 Mean :5.473e-04
## 3rd Qu.:6.641e-04 3rd Qu.:6.468e-04 3rd Qu.:6.391e-04
## Max. :2.285e-03 Max. :2.324e-03 Max. :2.348e-03
##
## shape45 shape46 shape47
## Min. :0.0000429 Min. :5.534e-05 Min. :4.349e-05
## 1st Qu.:0.0003457 1st Qu.:3.323e-04 1st Qu.:3.210e-04
## Median :0.0004633 Median :4.465e-04 Median :4.421e-04
## Mean :0.0005448 Mean :5.430e-04 Mean :5.431e-04
## 3rd Qu.:0.0006425 3rd Qu.:6.459e-04 3rd Qu.:6.475e-04
## Max. :0.0024146 Max. :2.378e-03 Max. :2.342e-03
##
## shape48 shape49 shape50
## Min. :3.953e-05 Min. :2.167e-05 Min. :6.256e-05
## 1st Qu.:3.128e-04 1st Qu.:3.114e-04 1st Qu.:3.153e-04
## Median :4.356e-04 Median :4.408e-04 Median :4.409e-04
## Mean :5.433e-04 Mean :5.467e-04 Mean :5.482e-04
## 3rd Qu.:6.631e-04 3rd Qu.:6.684e-04 3rd Qu.:6.736e-04
## Max. :2.458e-03 Max. :2.315e-03 Max. :2.259e-03
##
## shape51 shape52 shape53
## Min. :9.222e-05 Min. :0.0001067 Min. :8.834e-05
## 1st Qu.:3.211e-04 1st Qu.:0.0003351 1st Qu.:3.519e-04
## Median :4.500e-04 Median :0.0004544 Median :4.707e-04
## Mean :5.479e-04 Mean :0.0005482 Mean :5.515e-04
## 3rd Qu.:6.706e-04 3rd Qu.:0.0006591 3rd Qu.:6.532e-04
## Max. :2.374e-03 Max. :0.0022932 Max. :2.166e-03
##
## shape54 shape55 shape56
## Min. :9.699e-05 Min. :0.0001084 Min. :6.926e-05
## 1st Qu.:3.668e-04 1st Qu.:0.0003825 1st Qu.:4.032e-04
## Median :4.984e-04 Median :0.0005194 Median :5.421e-04
## Mean :5.557e-04 Mean :0.0005620 Mean :5.691e-04
## 3rd Qu.:6.546e-04 3rd Qu.:0.0006630 3rd Qu.:6.734e-04
## Max. :2.130e-03 Max. :0.0022557 Max. :2.132e-03
##
## shape57 shape58 shape59
## Min. :5.509e-05 Min. :3.757e-05 Min. :0.0000396
## 1st Qu.:4.123e-04 1st Qu.:4.268e-04 1st Qu.:0.0004392
## Median :5.608e-04 Median :5.868e-04 Median :0.0006083
## Mean :5.766e-04 Mean :5.884e-04 Mean :0.0006043
## 3rd Qu.:6.886e-04 3rd Qu.:7.095e-04 3rd Qu.:0.0007369
## Max. :1.990e-03 Max. :1.861e-03 Max. :0.0017404
##
## shape60 shape61 shape62
## Min. :4.163e-05 Min. :8.676e-05 Min. :0.0001340
## 1st Qu.:4.538e-04 1st Qu.:4.623e-04 1st Qu.:0.0004755
## Median :6.296e-04 Median :6.472e-04 Median :0.0006668
## Mean :6.236e-04 Mean :6.454e-04 Mean :0.0006714
## 3rd Qu.:7.737e-04 3rd Qu.:8.080e-04 3rd Qu.:0.0008439
## Max. :1.760e-03 Max. :1.922e-03 Max. :0.0020901
##
## shape63 shape64
## Min. :0.0001685 Min. :0.0001659
## 1st Qu.:0.0004985 1st Qu.:0.0005112
## Median :0.0006884 Median :0.0007058
## Mean :0.0007009 Mean :0.0007289
## 3rd Qu.:0.0008825 3rd Qu.:0.0009208
## Max. :0.0022627 Max. :0.0024313
## summary(texture_data)## species texture1 texture2
## Acer Capillipes: 16 Min. :0.000000 Min. :0.000000
## Acer Circinatum: 16 1st Qu.:0.000000 1st Qu.:0.000977
## Acer Mono : 16 Median :0.006836 Median :0.004883
## Acer Opalus : 16 Mean :0.021376 Mean :0.011693
## Acer Palmatum : 16 3rd Qu.:0.021972 3rd Qu.:0.014648
## Acer Pictum : 16 Max. :0.413090 Max. :0.147460
## (Other) :1503
## texture3 texture4 texture5
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.005859 Median :0.005859 Median :0.006836
## Mean :0.010230 Mean :0.015315 Mean :0.026646
## 3rd Qu.:0.014648 3rd Qu.:0.021484 3rd Qu.:0.044922
## Max. :0.118160 Max. :0.161130 Max. :0.201170
##
## texture6 texture7 texture8
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.001953 1st Qu.:0.000000
## Median :0.000977 Median :0.009766 Median :0.008789
## Mean :0.009449 Mean :0.016449 Mean :0.019495
## 3rd Qu.:0.008789 3rd Qu.:0.022461 3rd Qu.:0.030273
## Max. :0.146480 Max. :0.183590 Max. :0.176760
##
## texture9 texture10 texture11
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000977 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.006836 Median :0.003906 Median :0.003906
## Mean :0.015011 Mean :0.019787 Mean :0.019310
## 3rd Qu.:0.020508 3rd Qu.:0.022461 3rd Qu.:0.025391
## Max. :0.196290 Max. :0.315430 Max. :0.285160
##
## texture12 texture13 texture14
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.000000 Median :0.004883 Median :0.006836
## Mean :0.027292 Mean :0.009721 Mean :0.012895
## 3rd Qu.:0.007812 3rd Qu.:0.014648 3rd Qu.:0.018555
## Max. :0.507810 Max. :0.111330 Max. :0.184570
##
## texture15 texture16 texture17
## Min. :0.00000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.00000 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.00000 Median :0.000000 Median :0.003906
## Mean :0.01272 Mean :0.004877 Mean :0.014521
## 3rd Qu.:0.00000 3rd Qu.:0.002930 3rd Qu.:0.017578
## Max. :0.85352 Max. :0.119140 Max. :0.305660
##
## texture18 texture19 texture20
## Min. :0.000000 Min. :0.00000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.00293 1st Qu.:0.002930
## Median :0.000000 Median :0.01367 Median :0.008789
## Mean :0.005795 Mean :0.02442 Mean :0.014532
## 3rd Qu.:0.004883 3rd Qu.:0.03564 3rd Qu.:0.020508
## Max. :0.208980 Max. :0.30078 Max. :0.134770
##
## texture21 texture22 texture23
## Min. :0.000000 Min. :0.00000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.00000 1st Qu.:0.001953
## Median :0.000000 Median :0.00293 Median :0.007812
## Mean :0.002899 Mean :0.01319 Mean :0.017030
## 3rd Qu.:0.000000 3rd Qu.:0.01758 3rd Qu.:0.022461
## Max. :0.097656 Max. :0.17773 Max. :0.195310
##
## texture24 texture25 texture26
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.001953 Median :0.003906 Median :0.004883
## Mean :0.010304 Mean :0.010450 Mean :0.023417
## 3rd Qu.:0.011719 3rd Qu.:0.014648 3rd Qu.:0.027344
## Max. :0.134770 Max. :0.153320 Max. :0.302730
##
## texture27 texture28 texture29
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.003906
## Median :0.006836 Median :0.001953 Median :0.012695
## Mean :0.021175 Mean :0.010016 Mean :0.021362
## 3rd Qu.:0.030273 3rd Qu.:0.010742 3rd Qu.:0.030273
## Max. :0.227540 Max. :0.170900 Max. :0.214840
##
## texture30 texture31 texture32
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000977 1st Qu.:0.004883 1st Qu.:0.000000
## Median :0.007812 Median :0.015625 Median :0.000000
## Mean :0.011489 Mean :0.024020 Mean :0.006121
## 3rd Qu.:0.016602 3rd Qu.:0.033203 3rd Qu.:0.002930
## Max. :0.085938 Max. :0.285160 Max. :0.128910
##
## texture33 texture34 texture35 texture36
## Min. :0.00000 Min. :0.00000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.00000 1st Qu.:0.00000 1st Qu.:0.000977 1st Qu.:0.000000
## Median :0.00000 Median :0.01074 Median :0.006836 Median :0.000000
## Mean :0.02210 Mean :0.02676 Mean :0.010475 Mean :0.003334
## 3rd Qu.:0.01172 3rd Qu.:0.03418 3rd Qu.:0.014648 3rd Qu.:0.000000
## Max. :0.46094 Max. :0.35352 Max. :0.085938 Max. :0.115230
##
## texture37 texture38 texture39
## Min. :0.00000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.00000 1st Qu.:0.000977 1st Qu.:0.000977
## Median :0.00000 Median :0.016602 Median :0.008789
## Mean :0.02091 Mean :0.023625 Mean :0.017261
## 3rd Qu.:0.01074 3rd Qu.:0.036133 3rd Qu.:0.026367
## Max. :0.55664 Max. :0.217770 Max. :0.147460
##
## texture40 texture41 texture42
## Min. :0.000000 Min. :0.00000 Min. :0.000000
## 1st Qu.:0.001953 1st Qu.:0.00000 1st Qu.:0.000000
## Median :0.009766 Median :0.00000 Median :0.000977
## Mean :0.019240 Mean :0.01573 Mean :0.006061
## 3rd Qu.:0.026367 3rd Qu.:0.01074 3rd Qu.:0.006836
## Max. :0.166020 Max. :0.41113 Max. :0.075195
##
## texture43 texture44 texture45
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000977 1st Qu.:0.000000 1st Qu.:0.001465
## Median :0.007812 Median :0.005859 Median :0.007812
## Mean :0.014154 Mean :0.026033 Mean :0.016452
## 3rd Qu.:0.020508 3rd Qu.:0.031250 3rd Qu.:0.022461
## Max. :0.176760 Max. :0.412110 Max. :0.137700
##
## texture46 texture47 texture48
## Min. :0.00000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.00000 1st Qu.:0.004883 1st Qu.:0.000000
## Median :0.01172 Median :0.015625 Median :0.000977
## Mean :0.02188 Mean :0.018734 Mean :0.016594
## 3rd Qu.:0.03516 3rd Qu.:0.029297 3rd Qu.:0.021972
## Max. :0.17578 Max. :0.098633 Max. :0.249020
##
## texture49 texture50 texture51
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000977 1st Qu.:0.001953 1st Qu.:0.000000
## Median :0.007812 Median :0.009766 Median :0.000000
## Mean :0.011642 Mean :0.019195 Mean :0.013255
## 3rd Qu.:0.017578 3rd Qu.:0.024414 3rd Qu.:0.008301
## Max. :0.114260 Max. :0.232420 Max. :0.390630
##
## texture52 texture53 texture54
## Min. :0.000000 Min. :0.000000 Min. :0.00000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.00293
## Median :0.000000 Median :0.004883 Median :0.01269
## Mean :0.007108 Mean :0.014275 Mean :0.02243
## 3rd Qu.:0.009766 3rd Qu.:0.020508 3rd Qu.:0.03320
## Max. :0.117190 Max. :0.165040 Max. :0.29297
##
## texture55 texture56 texture57
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.000977
## Median :0.004883 Median :0.000000 Median :0.005859
## Mean :0.036817 Mean :0.005311 Mean :0.015576
## 3rd Qu.:0.045410 3rd Qu.:0.000000 3rd Qu.:0.020508
## Max. :0.429690 Max. :0.441410 Max. :0.172850
##
## texture58 texture59 texture60
## Min. :0.000000 Min. :0.000000 Min. :0.00000
## 1st Qu.:0.000000 1st Qu.:0.004883 1st Qu.:0.00000
## Median :0.000977 Median :0.012695 Median :0.00000
## Mean :0.011543 Mean :0.015980 Mean :0.01285
## 3rd Qu.:0.009766 3rd Qu.:0.021484 3rd Qu.:0.00000
## Max. :0.200200 Max. :0.106450 Max. :0.60645
##
## texture61 texture62 texture63
## Min. :0.000000 Min. :0.000000 Min. :0.000000
## 1st Qu.:0.000000 1st Qu.:0.000000 1st Qu.:0.000000
## Median :0.000000 Median :0.003906 Median :0.002930
## Mean :0.002637 Mean :0.019987 Mean :0.009054
## 3rd Qu.:0.000000 3rd Qu.:0.022461 3rd Qu.:0.013184
## Max. :0.151370 Max. :0.375980 Max. :0.086914
##
## texture64
## Min. :0.000000
## 1st Qu.:0.000977
## Median :0.012695
## Mean :0.019982
## 3rd Qu.:0.031250
## Max. :0.149410
## Boxplot can be used to qualitatively understand if there is separation of values range across species for each specific feature. Herein we do such analysis for a few features (due to limit on publishable post size). We define an helper function for the purpose.
species_boxplot <- function(dataset, variable) {
p <- ggplot(data = dataset, aes(x = species, y = eval(parse(text=variable)), col= species)) + theme(legend.position = "none") + geom_boxplot() + ylab(parse(text=variable))
p <- p + theme(axis.text.x = element_text(angle = 90, hjust = 1)) + ggtitle(paste(variable, "among species", sep = " "))
p
}species_boxplot(margin_data, "margin1")
species_boxplot(shape_data, "shape20")
species_boxplot(texture_data, "texture30")
Visual inspection of above boxplot suggests we have available potential features to help in prediction, as range of values tends to be different across species.
We then delete the 16th row from margin and shape datasets congruently with having a missing entry for texture dataset at the same row number.
margin_data <- margin_data[-16,]
shape_data <- shape_data[-16,]Let us add an identifier as associated to each row for all three datasets. That identifier eases table join operations.
margin_data <- mutate(margin_data, id = 1:nrow(margin_data))
shape_data <- mutate(shape_data, id = 1:nrow(shape_data))
texture_data <- mutate(texture_data, id = 1:nrow(texture_data))shape_data_2 <- shape_data
shape_data_2$species <- NULL
texture_data_2 <- texture_data
texture_data_2$species <- NULL
leaves_data <- margin_data
leaves_data <- left_join(leaves_data, shape_data_2, by = "id")
leaves_data <- left_join(leaves_data, texture_data_2, by = "id")
dim(leaves_data)## [1] 1599 194Saving the current enviroment for further analysis that will be outlined in my next post.
save.image(file='PlantLeafEnvironment.RData')References
[1] Charles Mallah, James Cope and James Orwell, "Plant leaf classification using probabilistic integration of shape, texture and margin features" [https://www.researchgate.net/publication/266632357_Plant_Leaf_Classification_using_Probabilistic_Integration_of_Shape_Texture_and_Margin_Features]
[2] 100 Plant Leaf Dataset [https://archive.ics.uci.edu/ml/datasets/One-hundred+plant+species+leaves+data+set]
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.