Orthogonal ratio of image processing (IoU)

This article fixed link: https://www.lao-wang.com/?p=114 reprint: Searky July 31, 2017 in Lao Wang's website published

Orthogonal ratio (INTERSECTION-OVER-UNION,IOU), a concept used in target detection, is the rate of overlap between the resulting candidate frame (candidate bound) and the original marker box (ground truth bound), i.e. the ratio of their intersection to the set. Ideally, the total overlap, which is the ratio of 1.

Calculation formula:

Python implementation code:

def Calculateiou (Candidatebound, Groundtruthbound):
    cx1 = candidatebound[0]
    cy1 = candidatebound[1]
    cx2 = CANDIDATEBOUND[2]
    cy2 = candidatebound[3]

    gx1 = groundtruthbound[0]
    gy1 = groundtruthbound[1]
    gx2 = GROUNDTRUTHBOUND[2]
    gy2 = groundtruthbound[3]

    Carea = (cx2-cx1) * (cy2-cy1) #C的面积
    Garea = (gx2-gx1) * (g Y2-GY1) #G的面积

    x1 = max (cx1, gx1)
    y1 = max (Cy1, gy1)
    x2 = min (cx2, gx2)
    y2 = min (cy2, gy2)
    w = m Ax (0, x2-x1)
    h = max (0, y2-y1)
    area = w * H #C ∩g

    IOU = areas/(Carea + garea-area) return

    IOU

def Calculateiou (Candidatebound, groundtruthbound): cx1 = candidatebound[0] Cy1 = candidatebound[1] cx2 = candidatebound[ 2] cy2 = candidatebound[3] gx1 = groundtruthbound[0] gy1 = groundtruthbound[1] gx2 = groundtruthbound[2] Gy2 = Groundtruth BOUND[3] Carea = (cx2-cx1) * (cy2-cy1) #C的面积 Garea = (gx2-gx1) * (gy2-gy1) #G的面积 x1 = max (cx1, gx1) y1 = max (Cy1, gy1) x2 = min (cx2, gx2) y2 = min (cy2, gy2) w = max (0, x2-x1) H = max (0, y2-y1) area = w * H #C ∩g size IOU = areas/(Care A + Garea-area) return IOU