Method and device for identifying state, electronic device and computer -readable storage medium

Patent metadata

Field	Value
Publication number	US-11763543-B2
Application number	US-202117562234-A
Country	US
Kind code	B2
Filing date	Dec 27, 2021
Priority date	Dec 15, 2021
Publication date	Sep 19, 2023
Grant date	Sep 19, 2023

Abstract

A method and device for identifying a state, electronic device and computer-readable storage medium are provided. The method includes: acquiring a to-be-detected image for a specific scene and determining a region of interest in the to-be-detected image; the region of interest being a region obtained by subtracting an occlusion range of the storage container in a closed state from an occlusion range in an open state; determining a positional relation between the region of interest and specific object regions where the specific objects are positioned in the to-be-detected image; determining at least one value based on values of pixels in the region of interest, when the positional relation represents that there is no specific object region partially overlapping the region of interest; and determining whether the storage container is in an open state or closed state based on the at least one value and a preset value range.

First claim

The invention claimed is:

1. A method for identifying a state, comprising:
acquiring a to-be-detected image for a specific scene, and determining a region of interest in the to-be-detected image, wherein the specific scene comprises a storage container configured to store specific objects; and the region of interest is a region obtained by subtracting an occlusion range of the storage container in a closed state from an occlusion range of the storage container in an open state;
determining a positional relation between the region of interest and specific object regions in which the specific objects are positioned in the to-be-detected image;
determining at least one value based on pixel values of pixels in the region of interest when the positional relation represents that there is no specific object region partially overlapping the region of interest, wherein each value represents a number of storage spaces in the storage container configured to place the specific objects; and
determining whether the storage container is in the open state or the closed state based on the at least one value and a preset value range.

2. The method of claim 1 , wherein the step of determining whether the storage container is in the open state or the closed state based on the at least one value and the preset value range comprises:
determining a target value from the at least one value; and
determining whether the storage container is in the open state or the closed state based on the target value and the preset value range.

3. The method of claim 2 , wherein the step of determining whether the storage container is in the open state or the closed state based on the target value and the preset value range comprises at least one of:
determining that the storage container is in the open state, when the target value falls within the preset value range; or
determining that the storage container is in the closed state, when the target value does not fall within the preset value range.

4. The method of claim 1 , wherein the step of determining the region of interest in the to-be-detected image comprises:
acquiring a preset standard image of the specific scene, wherein the preset standard image comprises a preset standard region;
determining a mapping relation between the preset standard image and the to-be-detected image; and
determining the region of interest mapped with the preset standard region in the to-be-detected image according to the mapping relation.

5. The method of claim 4 , wherein the step of determining the region of interest mapped with the preset standard region in the to-be-detected image according to the mapping relation comprises:
determining multiple mapped boundary coordinate vertices according to the mapping relation and multiple boundary coordinate vertices of the preset standard region; and
determining the region of interest according to a region determined from the mapped boundary coordinate vertices.

6. The method of claim 2 , wherein the step of determining the target value from the at least one value comprises:
determining an average value based on the at least one value, and determining the average value as the target value.

7. The method of claim 1 , wherein the step of determining the at least one value based on the pixel values of the pixels in the region of interest comprises:
determining a bounding rectangle region corresponding to the region of interest;
increasing or decreasing a number of the pixel values of pixels in each horizontal direction of at least one horizontal direction in the bounding rectangle region to a specific number to obtain a pixel value array corresponding to the horizontal direction, wherein each horizontal direction is represented by a horizontal line having a preset width in the region of interest;
performing Fast Fourier Transform on the pixel value array corresponding to the horizontal direction to obtain at least one frequency value corresponding to each horizontal direction; and
determining a maximum frequency value from the at least one frequency value, and determining the maximum frequency value corresponding to each horizontal direction of the at least one horizontal direction as the at least one value.

8. The method of claim 1 , wherein the step of determining the positional relation between the region of interest and the specific object regions in which the specific objects are positioned in the to-be-detected image comprises at least one of:
determining a dimension scale of the specific object region when the region of interest of the to-be-detected image comprises the specific objects; and determining that the positional relation represents that there is no specific object region partially overlapping the region of interest when the dimension scale is smaller than or equal to a preset scale; or
determining the positional relation represents that there is no specific object region partially overlapping the region of interest when the region of interest does not comprise the specific objects.

9. The method of claim 8 , wherein, after determining the dimension scale of each specific object region, the method further comprises:
determining that the positional relation represents that there is a specific object region at least partially overlapping the region of interest when the dimension scale of the specific object region is larger than the preset scale and the specific object region overlaps the region of interest; and
determining that the storage container is in the open state when the positional relation represents that there is a specific object region at least partially overlapping the region of interest.

10. The method of claim 9 , further comprising:
when the to-be-detected image comprises the specific objects, and at least one vertex on any midline segment of the specific object region is positioned outside the region of interest, and at least one point on any midline segment is positioned within the region of interest, determining that the specific object region overlaps the region of interest.

11. A device for identifying a state, comprising:
a memory for storing instructions; and
a processor,
wherein the processor is configured to execute the instructions to:
acquire a to-be-detected image for a specific scene; and
determine a region of interest in the to-be-detected image, wherein the specific scene comprises a storage container configured to store specific objects, and the region of interest is a region obtained by subtracting an occlusion range of the storage container in a closed state from an occlusion range of the storage container in an open state; determine a positional relation between the region of interest and specific object regions in which the specific objects are positioned in the to-be-detected image; determine at least one value based on pixel values of pixels in the region of interest when the positional relation represents that there is no specific object region partially overlapping the region of interest, wherein each value represents a number of storage spaces in the storage container configured to place the specific objects; and determine whether the storage container is in the open state or the closed state based on the at least one value and a preset value range.

12. The device of claim 11 , wherein the processor is configured to execute the instructions to:
determine a target value from the at least one value; and
determine whether the storage container is in the open state or the closed state based on the target value and the preset value range.

13. The device of claim 12 , wherein the processor is configured to execute the instructions

Assignees

• Sensetime Int Pte Ltd

Inventors

• Zhang Wenbin
• Zhang Yao
• Zhang Shuai
• Yi Shuai

Classifications

• G06V20/50 Primary

  Context or environment of the image · CPC title

• G06V20/52 Primary

  Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) · CPC title

• G06V10/273 Primary

  removing elements interfering with the pattern to be recognised · CPC title

• G06V10/25

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06V10/48

  by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation · CPC title