Steel size and quantity identification method based on deep learning, intelligent equipment and storage medium

Abstract

The invention discloses a steel size and quantity identification method based on deep learning, intelligent equipment and a storage medium. The identification method comprises the following steps of:shooting a plurality of steel pile end face images through a camera, collecting camera aperture values and focal length values corresponding to the steel pile end face images, and performing trainingto obtain a steel number detection and size recognition neural network model by utilizing the plurality of steel pile end face images and the aperture values and focal length values of the cameras corresponding to the steel pile end face images; and shooting an end face image of the steel pile to be identified by using a camera, inputting the end face image of the steel pile to be identified and the aperture value and the focal length value of the corresponding camera into the model, and outputting the steel quantity and size information of the image by the model. Thus, the number, coordinatesand sizes of the steel in the steel pile can be rapidly obtained only by collecting the end face image of the steel pile and the aperture and focal length values of the camera through the camera, theefficiency and accuracy of steel counting and size recognition are effectively improved, and the labor cost and time cost are greatly reduced.

Description

technical field [0001] The invention relates to the field of image vision technology, in particular to a deep learning-based recognition method for steel size and quantity, an intelligent device and a storage medium. Background technique [0002] In the links of steel production, transportation, sales, etc., whether it is storage or sale, whether it is for steel companies, steel sellers or steel buyers, in order to reduce possible economic risks and disputes, each link must accurately calculate the steel At the same time, due to the different sizes, thicknesses and uses of various types of steel, it is also necessary to measure the size of the steel. [0003] At present, there are the following methods for controlling the quantity and size of steel products: [0004] 1. Manual counting and measurement. [0005] Manual marking and measurement are the most commonly used methods at this stage. Workers use pens of different colors to mark the steel one by one. This traditiona...

AI Technical Summary

Problems solved by technology

This method has the following disadvantages: 1) only a single measurement can be performed, and the efficiency is low; 2) in harsh environments, the sensitivity of the photoelectric cell will be reduced, and the error is large; 3) the size of the steel cannot be measured at the same time

However, these methods at this stage have the following problems: 1) For traditional algorithms, the actual storage environment of steel is complex and changeable, and light changes, steel occlusion, etc. are prone to occur, and traditional algorithms have high requirements for scenes. The practicality of the algorithm is not very strong; 2) For some deep learning algorithms at the present stage, although the model has a high theoretical detection rate due to the diversity of scenes and the incompleteness of data, the actual use The accuracy and recall rate of steel number detection are not high

At the same time, the diameter of the steel varies widely (many types between 12-32), and the interface is irregular in shape and color, and the shooting angle and distance are not controlled, which leads to the detection of the algorithm model in the process of actual use. The effect is difficult to stabilize; 3) In the current method, there is no method that can simultaneously realize the quantitative statistics of steel and classify the size level of steel bundles

Images