% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/sharkDetectoR.R
\name{detect_and_classify}
\alias{detect_and_classify}
\title{Submit a single image for combined detection + hierarchical classification (client-side)}
\usage{
detect_and_classify(
  image_path,
  threshold = 0.25,
  draw_boxes = FALSE,
  crop = FALSE,
  topk = 1,
  save_dir = "./"
)
}
\arguments{
\item{image_path}{Path to a local image file (JPEG/PNG).}

\item{threshold}{Numeric in [0,1] controlling YOLO‐confidence threshold. Default = 0.25.}

\item{draw_boxes}{Logical; if TRUE, an annotated image with drawn boxes is downloaded.}

\item{crop}{Logical; if TRUE, each detected shark patch (cropped) is downloaded.}

\item{topk}{Integer ≥ 1: how many taxonomy candidates to return per detection. Default = 1.}

\item{save_dir}{Character; directory where downloaded images are saved. Default = `"./"`.
A subfolder `"images"` will be created if it does not exist.}
}
\value{
A data.frame with one row per detection × per candidate, with columns:
  \itemize{
    \item \code{xmin, ymin, xmax, ymax} – bounding‐box coordinates (numeric)
    \item \code{score}   – detection confidence (numeric)
    \item \code{label}   – `"shark"` (character) or integer label otherwise
    \item \code{order, family, genus, species} (character)
    \item \code{p_order, p_family, p_genus, p_species, joint_score} (numeric)
    \item \code{annotated_image} – local filename of the annotated image (if \code{draw_boxes=TRUE}), or NA
    \item \code{cropped_image} – local filename of the cropped patch (if \code{crop=TRUE}), or NA
  }
}
\description{
Calls the Flask API endpoint `/detect_and_classify` at \code{http://sharkpulse.cnre.vt.edu}.
Optionally downloads any annotated / cropped images into `save_dir/images/`.
Returns a data frame with one row per detected bounding‐box and per taxonomy candidate.
}
\examples{
\dontrun{
  # topk = 1: one row per detection, download annotated box
  df1 <- detect_and_classify(
    image_path = "shark.jpg",
    threshold  = 0.3,
    draw_boxes = TRUE,
    crop       = FALSE,
    topk       = 1,
    save_dir   = "./results"
  )
  print(df1)

  # topk = 3: three rows per detection, download crops
  df3 <- detect_and_classify(
    image_path = "shark.jpg",
    threshold  = 0.3,
    draw_boxes = FALSE,
    crop       = TRUE,
    topk       = 3,
    save_dir   = "./results"
  )
  print(df3)
}
}