US-12626387-B2 - Method and system for automatically determining dimensions of a carton box

US12626387B2US 12626387 B2US12626387 B2US 12626387B2US-12626387-B2

Abstract

Embodiments of present disclosure relates to method and dimension determination system for determining dimensions of carton box. The dimension determination system receives plurality of datapoints associated with open carton box. The dimension determination system extracts depth datapoints from plurality of datapoints for identifying plurality of flaps. The dimension determination system determines height of open carton box using depth datapoints. The dimension determination system generates contour and estimates rectangle figure for contour of open carton box. Further, the dimension determination system determines width and length by extracting vertices of rectangle figure. Thereafter, the dimension determination system utilises width, length and height to obtain dimensions of open carton box. Thus, the present disclosure automatically determines dimensions of carton box.

Inventors

Navya VEPAKOMMA
Yadhunandan Ullam Subbaraya

Assignees

WIPRO LIMITED

Dates

Publication Date: 20260512
Application Date: 20231011
Priority Date: 20230829

Claims (15)

1 . A method of determining dimensions of a carton box, the method comprising: receiving, by a processor of a dimension determination system, a plurality of data points associated with an open carton box placed on a surface using one or more depth sensors, wherein the one or more depth sensors are located vertically above the surface; extracting, by the processor of the dimension determination system, depth data points from the plurality of data points, wherein the depth data points are clustered based on closeness between the depth data points using a clustering technique to identify a plurality of flaps for the open carton box, and wherein the plurality of flaps of the open carton box is represented as three-dimensional coordinate points; determining, by the processor of the dimension determination system, a height of the open carton box by determining a cross section of the open carton box using the depth data points, wherein the height is determined based on a difference between a highest point of the open carton box, representing a rim of the open carton box, and a lowest point of the open carton box obtained from a central area of the open carton box, using the cross section, and wherein the cross section of the open carton box is determined by segmenting the depth data points based on the highest point of the open carton box; generating, by the processor of the dimension determination system, a contour for the open carton box by converting the three-dimensional coordinate points of the plurality of flaps into a two-dimensional image based on one or more parameters of the one or more depth sensors; estimating, by the processor of the dimension determination system, a minimum bounding rectangle figure for the contour of the open carton box, wherein the minimum bounding rectangle figure encloses the contour and is represented as a two-dimensional image; and determining, by the processor of the dimension determination system, a width and a length for the open carton box by extracting vertices of the minimum bounding rectangle figure, wherein the width, the length and the height of the open carton box is used to obtain the dimensions of the open carton box.
2 . The method as claimed in claim 1 , wherein the closeness between the depth data points is determined based on a depth difference and a distance between the depth data points.
3 . The method as claimed in claim 1 and further comprising: validating the depth data points of the plurality of flaps by: detecting, by the processor of the dimension determination system, a plurality of corner points of the open carton box using a trained Neural Network (NN) model; and comparing, by the processor of the dimension determination system, the plurality of corner points of the open carton box with the three-dimensional coordinate points of the plurality of flaps.
4 . The method as claimed in claim 1 , wherein extracting the vertices of the minimum bounding rectangle figure comprises: converting, by the processor of the dimension determination system, the two-dimensional image of the minimum bounding rectangle figure into three-dimensional coordinate points of the minimum bounding rectangle figure based on the one or more parameters of the one or more depth sensors.
5 . The method as claimed in claim 1 , wherein the one or more parameters comprises focal length, aperture, field-of-view, and resolution of the one or more depth sensors.
6 . The method as claimed in claim 1 , wherein the two-dimensional image of the minimum bounding rectangle figure is determined using one or more fitting techniques.
7 . The method as claimed in claim 1 , wherein receiving the plurality of data points further comprises: calibrating, by the processor of the dimension determination system, the surface on which the open carton box is placed using sensor data obtained from the one or more depth sensors, wherein calibration is performed to remove the depth data points from the plurality of data points, associated with the surface.
8 . A dimension determination system for determining dimensions of a carton box, comprising: a processor; and a memory communicatively coupled to the processor, wherein the memory stores processor-executable instructions, which, on execution, cause the processor to: receive a plurality of data points associated with an open carton box placed on a surface using one or more depth sensors, wherein the one or more depth sensors are located vertically above the surface; extract depth data points from the plurality of data points, wherein the depth data points are clustered based on closeness between the depth data points using a clustering technique to identify a plurality of flaps for the open carton box, and wherein the plurality of flaps of the open carton box is represented as three-dimensional coordinate points; determine a height of the open carton box by determining a cross section of the open carton box using the depth data points, wherein the height is determined based on a difference between a highest point of the open carton box, representing a rim of the open carton box, and a lowest point of the open carton box obtained from a central area of the open carton box, using the cross section, and wherein the cross section of the open carton box is determined by segmenting the depth data points based on the highest point of the open carton box; generate a contour for the open carton box by converting the three-dimensional coordinate points of the plurality of flaps into a two-dimensional image based on one or more parameters of the one or more depth sensors; estimate a minimum bounding rectangle figure for the contour of the open carton box, wherein the minimum bounding rectangle figure encloses the contour and is represented as a two-dimensional image, and wherein the two-dimensional image of the minimum bounding rectangle figure is determined using one or more fitting techniques; and determine a width and a length for the open carton box by extracting vertices of the minimum bounding rectangle figure, wherein the width, the length and the height of the open carton box is used to obtain the dimensions of the open carton box.
9 . The dimension determination system as claimed in claim 8 , wherein the closeness between the depth data points is determined based on a depth difference and a distance between the depth data points.
10 . The dimension determination system as claimed in claim 8 , wherein the processor is configured to: validate the depth data points of the plurality of flaps by: detecting a plurality of corner points of the open carton box using a trained Neural Network (NN) model; and comparing the plurality of corner points of the open carton box with the three-dimensional coordinate points of the plurality of flaps.
11 . The dimension determination system as claimed in claim 8 , wherein the processor extracts the vertices of the minimum bounding rectangle figure by: converting the two-dimensional image of the minimum bounding rectangle figure into three-dimensional coordinate points of the minimum bounding rectangle figure based on the one or more parameters of the one or more depth sensors.
12 . The dimension determination system as claimed in claim 8 , wherein the one or more parameters comprises focal length, aperture, field of view, and resolution of the one or more depth sensors.
13 . The dimension determination system as claimed in claim 8 , wherein the two-dimensional image of the minimum bounding rectangle figure is determined using one or more fitting techniques.
14 . The dimension determination system as claimed in claim 8 , wherein the processor receives the plurality of data points by: calibrating the surface on which the open carton box is placed using sensor data obtained from one or more sensors, wherein calibration is performed to remove the depth data points associated with the surface.
15 . A non-transitory computer readable medium including instruction stored thereon that when processed by at least one processor cause a dimension determination system to perform operation comprising: receiving a plurality of data points associated with an open carton box placed on a surface using one or more depth sensors, wherein the one or more depth sensors are located vertically above the surface; extracting depth data points from the plurality of data points, wherein the depth data points are clustered based on closeness between the depth data points using a clustering technique to identify a plurality of flaps for the open carton box, and wherein the plurality of flaps of the open carton box is represented as three-dimensional coordinate points; determining a height of the open carton box by determining a cross section of the open carton box using the depth data points, wherein the height is determined based on a difference between a highest point of the open carton box, representing a rim of the open carton box, and a lowest point of the open carton box obtained from a central area of the open carton box, using the cross section, and wherein the cross section of the open carton box is determined by segmenting the depth data points based on the highest point of the open carton box; generating a contour for the open carton box by converting the three-dimensional coordinate points of the plurality of flaps into a two-dimensional image based on one or more parameters of the one or more depth sensors; estimating a minimum bounding rectangle figure for the contour of the open carton box, wherein the minimum bounding rectangle figure encloses the contour and is represented as a two-dimensional image; and determining a width and a length for the open carton box by extracting vertices of the minimum bounding rectangle figure, wherein the width, the length and the height of the open carton box is used to obtain the dimensions of the open carton box.

Description

TECHNICAL FIELD The present subject matter is related in general to packing and logistics industry, more particularly, but not exclusively, the present subject matter relates to a method and system for determining dimensions of a carton box. BACKGROUND Generally, warehouse, packaging, and logistics domains deal with carton boxes. Various items which come on a conveyor belt needs to be packed in an efficient manner into the carton boxes. For this purpose, the carton boxes are measured, and the items are packed into the carton boxes based on a volume and dimensions of the carton boxes. However, taking measurements of various carton boxes manually can be extremely time consuming and prone to errors. Currently, most of the existing systems are used to compute the dimensions of closed carton boxes. Particularly, the dimensions of the closed carton boxes which are present on the conveyor belt are computed by fitting minimum bounding cuboid or by reading machine readable tags which are present on packages. While, in other existing systems, when a closed carton box enters a Region of Interest (ROI), the dimensions of the closed carton box is computed. However, the existing systems are limited to conveyor belt system and do not compute dimensions of an open carton box. Also, the existing systems are limited to computing the dimensions of a single closed carton box at a time and do not support computation of dimensions of multiple open carton boxes in same scene. The information disclosed in this background of the disclosure section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person skilled in the art. SUMMARY In an embodiment, the present disclosure relates to a method of determining dimensions of a carton box. The method comprises receiving a plurality of data points associated with an open carton box placed on a surface using one or more depth sensors. The one or more depth sensors are located vertically above the surface. Upon receiving, the method comprises extracting depth data points from the plurality of data points. The depth data points are clustered based on closeness between the depth data points using a clustering technique to identify a plurality of flaps for the open carton box. The plurality of flaps of the open carton box is represented as three-dimensional coordinate points. Upon extracting, the method comprises determining a height of the open carton box by determining a cross section of the open carton box using the depth data points. Upon determining, the method comprises generating a contour for the open carton box by converting the three-dimensional coordinate points of the plurality of flaps into a two-dimensional image based on one or more parameters of the one or more depth sensors. Upon generating, the method comprises estimating a minimum bounding rectangle figure for the contour of the open carton box. The minimum bounding rectangle figure encloses the contour and is represented as a two-dimensional image. Thereafter, the method comprises determining a width, and a length for the open carton box by extracting vertices of the minimum bounding rectangle figure. The width, the length and the height of the open carton box is used to obtain the dimensions of the open carton box. In an embodiment, the present disclosure relates to a dimension determination system for determining dimensions of a carton box. The dimension determination system includes a processor and a memory communicatively coupled to the processor. The memory stores processor-executable instructions, which on execution cause the processor to determine dimensions of a carton box. The dimension determination system receives a plurality of data points associated with an open carton box placed on a surface using one or more depth sensors. The one or more depth sensors are located vertically above the surface. Upon receiving, the dimension determination system extracts depth data points from the plurality of data points. The depth data points are clustered based on closeness between the depth data points using a clustering technique to identify a plurality of flaps for the open carton box. The plurality of flaps of the open carton box is represented as three-dimensional coordinate points. Upon extracting, the dimension determination system determines a height of the open carton box by determining a cross section of the open carton box using the depth data points. Upon determining, the dimension determination system generates a contour for the open carton box by converting the three-dimensional coordinate points of the plurality of flaps into a two-dimensional image based on one or more parameters of the one or more depth sensors. Upon generating, the dimension determination system estimates a minimum bounding rectangle figure for the contour of the open carton bo