A window counting method and device based on multi-category incremental learning

Abstract

The invention provides a window-based counting method and device based on multi-category incremental learning, which makes full use of the characteristics of regular target shapes, establishes a direct relationship between pixels and numbers, reduces cumulative errors, and realizes fast counting, while allowing objects Sticking, ignoring non-uniform density and mass factors. In order to avoid the limitation of the shooting angle when collecting images, there will be errors in the pixels of each area of ​​the target image, the detection range is divided into partitions by windows, and each window is counted separately to reduce the error. The device uses a dark box to avoid the interference of external light sources, and has very low requirements on the external environment; the multi-window template partitions the counting area to avoid the rolling of target objects and the adhesion between adjacent windows. The invention realizes high-accuracy fast counting of regular objects with light weight, small volume and uneven density similar to industrial sealing felt pads.

Description

technical field [0001] The invention relates to image processing and machine learning technology, in particular to fast counting of regular objects with light weight, small volume and uneven density. Background technique [0002] Industrial sealing felt pad is a sealing gasket made by punching, which has the functions of sealing, heat insulation, sound insulation, shockproof, filtering, etc. It is widely used in various industrial fields such as home appliances, musical instruments, sports equipment, automobiles, and cultural products. , Accurately calculating the number of felt pads is very important to ensure the economic benefits of the manufacturer. [0003] The basic characteristics of felt pads are small size, light weight, and regular shape. Since the processed raw materials contain different types of wool, the final finished product will have uneven density and thickness due to the length and thickness of the wool, resulting in errors. The traditional methods are ge...

AI Technical Summary

Problems solved by technology

The weighing method requires that the weight of the weighing object is basically the same and meets the minimum unit of measurement, but the felt pad is small in size and light in weight, and uneven density and thickness are likely to cause large errors

The manual counting method is relatively slow, inefficient, and easily affected by human factors, occupying and wasting human resources

[0004] How to improve the efficiency and accuracy of counting regular objects with light weight, small volume and uneven density similar to industrial sealing felt pads is an urgent problem to be solved

Images