Conveyor belt static package detection method and system based on depth camera

Abstract

The invention belongs to the technical field of measurement. The invention discloses a conveyor belt static package detection method and a system based on a depth camera. The method comprises the steps of giving a fixed detection region D, comparing the depth images of two adjacent frames; counting the number N of points with obviously changed gray values in the fixed area D, setting a threshold value for judging whether a moving object exists or a proportion threshold value for static detection, and judging that the moving object exists when the number of the change points exceeds the threshold value for judging whether the moving object exists or the proportion of the change points exceeds the proportion threshold value for corresponding static detection; when and only when the moving object exists, the volume measurement method is used for wrapping. The method and the system can automatically work, are simple to operate, are convenient to implement, do not need manual judgment, arehigh in accuracy, and greatly improve the efficiency. The method can be nested in other systems to provide accurate and reliable services for other systems.

Description

technical field [0001] The invention belongs to the technical field of measurement, and in particular relates to a method and system for detecting static parcels on a conveyor belt based on a depth camera. Background technique [0002] At present, the commonly used existing technologies in the industry are as follows: as e-commerce increasingly penetrates into people's daily life, the logistics industry develops rapidly. Every day, various items are transferred by express companies, and the pressure on express companies to sort items is also increasing. When sorting items, the courier company must first perform operations such as security inspection, weighing, measuring, and classification of the items, and then transfer them according to the type and weight of the items. In the above process, hand-held detectors are generally used for manual security inspection, and then the package is placed on the weighing platform for load-bearing to obtain the weight of the item, and t...

AI Technical Summary

Problems solved by technology

And when loading and unloading goods, it often happens that the package stays on the conveyor belt for a long time, or the package can also be detected by the flow volume measurement algorithm, which makes some volume measurement systems accumulate wrong results, resulting in a decrease in accuracy

[0003] However, the use of hand-held detectors for detection is not only complicated, inefficient, and consumes a lot of manpower and material resources, but also it is easy to make mistakes when the identification code information corresponds to other information; and the flow volume measurement algorithm, when the object is still under the camera, cannot If detected in time, the volume will be calculated all the time, causing obvious errors; and due to the influence of deep image noise, even when there is no goods on the conveyor belt, the volume will change slightly, and the long-term accumulation of errors cannot be ignored

[0004] To sum up, the problems existing in the existing technology are: the use of hand-held detectors for detection is not only complicated in process, low in efficiency, but also consumes a lot of manpower and material resources, and it is easy to make mistakes when barcode information corresponds to other information (such as package volume, etc.); The flow volume measurement algorithm, when detecting stationary objects, will always acquire the volume, so that when the conveyor belt has no packages, it will also accumulate small errors, and the accumulated errors will be large after long-term operation

Due to the imaging characteristics of 3D cameras, objects of different sizes have a certain relationship with the amount of noise generated, so it is difficult to determine a fixed threshold

Images