^j0ddlmZddlmZddlmZddlZddlmZ ddl m Z ddl m Z ddlmZerddlmZd ZGd d eZ dd Zdd Z ddZddZej0f ddZy)) annotations)Enum) TYPE_CHECKINGN)ORIENTED_BOX_COORDINATES) CompactMask) MetricTarget) Detections)ii$c eZdZdZdZdZdZdZy)ObjectSizeCategorya Enum for object size categories based on area in pixels. Small: area < 32^2 Medium: 32^2 <= area < 96^2 Large: area >= 96^2 Example: ```pycon >>> from supervision.metrics.utils.object_size import ObjectSizeCategory >>> ObjectSizeCategory.SMALL.value 1 >>> ObjectSizeCategory.MEDIUM.value 2 >>> ObjectSizeCategory.LARGE.value 3 ``` N)__name__ __module__ __qualname____doc__ANYSMALLMEDIUMLARGEp/var/www/ramen.bs-engineer-server.com/venv/lib/python3.12/site-packages/supervision/metrics/utils/object_size.pyr r s( C E F Err c|tjk(r t|S|tjk(r t |S|tj k(r t |Std)aq Get the size category of an object. Distinguish based on the metric target. Args: data: The object data, shaped (N, ...). metric_target: Determines whether boxes, masks or oriented bounding boxes are used. Returns: The size category of each object, matching the enum values of ObjectSizeCategory. Shaped (N,). Example: ```pycon >>> import numpy as np >>> from supervision.metrics.core import MetricTarget >>> from supervision.metrics.utils.object_size import get_object_size_category >>> xyxy = np.array([ ... [0, 0, 10, 10], # 100 (Small) ... [0, 0, 50, 50], # 2500 (Medium) ... [0, 0, 100, 100] # 10000 (Large) ... ]) >>> get_object_size_category(xyxy, MetricTarget.BOXES) array([1, 2, 3]) ``` Invalid metric type)rBOXESget_bbox_size_categoryMASKSget_mask_size_categoryORIENTED_BOUNDING_BOXESget_obb_size_category ValueError)data metric_targets rget_object_size_categoryr&.s[< ***%d++ ***%d++ <<<$T** * ++rct|jdk7s|jddk7r td|dddf|dddfz }|dddf|dddfz }||z}tj|jt j j}t\}}t jj|||k<t jj|||k\||kz<t jj|||k\<|S)a Get the size category of a bounding boxes array. Args: xyxy: The bounding boxes array shaped (N, 4). Returns: The size category of each bounding box, matching the enum values of ObjectSizeCategory. Shaped (N,). Example: ```pycon >>> import numpy as np >>> from supervision.metrics.utils.object_size import get_bbox_size_category >>> xyxy = np.array([ ... [0, 0, 31, 31], # 961 (Small) ... [0, 0, 32, 32], # 1024 (Medium) ... [0, 0, 95, 95], # 9025 (Medium) ... [0, 0, 96, 96] # 9216 (Large) ... ]) >>> get_bbox_size_category(xyxy) array([1, 2, 2, 3]) ``` rr z$Bounding boxes must be shaped (N, 4)Nrr) lenshaper#npfullr rvalueSIZE_THRESHOLDSrrr)xyxywidthheightareasresultSMLGs rrrUs4 4::!tzz!}1?@@ AJad #E !Q$Z$q!t* $F FNE WWU[["4"8"8">"> ?F FB+1177F52:+=+D+D+J+JFERKEBJ '(,2288F5B; Mrct|tr |j}n:t|jdk7r t dt j|d}t j|jtjj}t\}}tjj|||k<tjj|||k\||kz<tjj|||k\<|S)a Get the size category of detection masks. Args: mask: The mask array shaped (N, H, W), or a :class:`~supervision.detection.compact_mask.CompactMask`. Returns: The size category of each mask, matching the enum values of ObjectSizeCategory. Shaped (N,). Example: ```pycon >>> import numpy as np >>> from supervision.metrics.utils.object_size import get_mask_size_category >>> mask = np.zeros((3, 100, 100), dtype=bool) >>> mask[0, 0:10, 0:10] = True # 100 (Small) >>> mask[1, 0:50, 0:50] = True # 2500 (Medium) >>> mask[2, 0:100, 0:100] = True # 10000 (Large) >>> get_mask_size_category(mask) array([1, 2, 3]) ``` rzMasks must be shaped (N, H, W))r r)axis) isinstancerarear)r*r#r+sumr,r rr-r.rrr)maskr2r3r4r5s rr r ~s6$ $  tzz?a => >t&) WWU[["4"8"8">"> ?F FB+1177F52:+=+D+D+J+JFERKEBJ '(,2288F5B; Mrct|jdk7s$|jddk7s|jddk7r td|dddddf}|dddddf}|j\}}}}|j\}}} } dt j ||z|| zz|| zz||zz||z||zz|| zz|| zzz z} t j | jtjj} t\} }tjj| | | k<tjj| | | k\| |kz<tjj| | |k\<| S) a Get the size category of a oriented bounding boxes array. Args: xyxyxyxy: The bounding boxes array shaped (N, 4, 2). Returns: The size category of each bounding box, matching the enum values of ObjectSizeCategory. Shaped (N,). Example: ```pycon >>> import numpy as np >>> from supervision.metrics.utils.object_size import get_obb_size_category >>> obb = np.array([ ... [[0, 0], [10, 0], [10, 10], [0, 10]], # 100 (Small) ... [[0, 0], [50, 0], [50, 50], [0, 50]], # 2500 (Medium) ... [[0, 0], [100, 0], [100, 100], [0, 100]] # 10000 (Large) ... ]) >>> get_obb_size_category(obb) array([1, 2, 3]) ``` rr r(rz0Oriented bounding boxes must be shaped (N, 4, 2)Nrg?)r)r*r#Tr+absr,r rr-r.rrr)xyxyxyxyxyx1x2x3x4y1y2y3y4r2r3r4r5s rr"r"sl2 8>>a8>>!#4#9X^^A=NRS=SKLL AqAAqASSNBBSSNBB "&& b27 R"W $rBw . 7R"W rBw &b 0 2 E WWU[["4"8"8">"> ?F FB+1177F52:+=+D+D+J+JFERKEBJ '(,2288F5B; Mrc|tjk(rt|jS|tjk(r,|j t dt|j S|tjk(rY|jjt t dttj|jtSt d)aX Get the size category of a detections object. Args: detections: The detections object. metric_target: Determines whether boxes, masks or oriented bounding boxes are used. Returns: The size category of each bounding box, matching the enum values of ObjectSizeCategory. Shaped (N,). z Detections mask is not availablez4Detections oriented bounding boxes are not availabler)rrrr/rr;r#r r!r$getrr"r+array) detectionsr%s rget_detection_size_categoryrNs ***%joo66 *** ?? "?@ @%joo66 <<< ??  7 8 @ST T$ HHZ__%=> ?   * ++r)r$z npt.NDArrayr%rreturnnpt.NDArray[np.int_])r/npt.NDArray[np.float32]rOrP)r;z#npt.NDArray[np.bool_] | CompactMaskrOrP)r?rQrOrP)rMr r%rrOrP) __future__renumrtypingrnumpyr+ numpy.typingnptsupervision.configr"supervision.detection.compact_maskrsupervision.metrics.corersupervision.detection.corer r.r r&rr r"rrNrrrr\s" 7:15 6$, $,&2$,$,N&R' -''T+^;G:L:L,,+7,,r