The invention discloses a method and device for measuring an icing thickness of a power transmission line. The method comprises the steps of collecting a four-eye image of the power transmission line;inputting the four-eye image of the power transmission line into a convolutional neural network model, and outputting a target parallax error point set of the power transmission line; calculating a depth point set of the power transmission line according to the target parallax error point set of the power transmission line; determining a three-dimensional model of the power transmission line according to the depth point set of the power transmission line; obtaining cross section information of a conducting wire of the power transmission line according to the three-dimensional model of the power transmission line; and determining an equivalent icing thickness of the power transmission line according to the cross section information of the conducting wire of the power transmission line. Byadopting the method for measuring the icing thickness of the power transmission line in the scheme, visual matching is performed on the four-eye image of the power transmission line based on the convolutional neural network model, the accuracy of measuring the equivalent icing thickness of the power transmission line is improved, and power companies can be effectively guided to perform deicing work.

Method and device for measuring icing thickness of power transmission line

The invention relates to the technical field of sorting bag equipment, in particular to a sorting device for classifying bags. The sorting device comprises a front conveying belt, a machine frame, a main controller, a scanning device and a rear conveying belt, wherein the scanning device comprises a first scanner, a second scanner, a third scanner, a fourth scanner and a fifth scanner which are correspondingly electrically connected with the main controller; the rear conveying belt is positioned at the output end of the front conveying belt; the rear conveying belt comprises a main channel, aplurality of branch channels and a plurality of control valves, one control valve is used for controlling the communication or isolation of the main channel and the branch channels, and the main channel is located at the output end of the front conveying belt, and each control valve is electrically connected with the main control device; and the front conveying belt, the scanning device and the rear conveying belt are arranged, the scanning device comprises five scanners, the five directions of the bags are scanned to obtain logistics information, and the main controller controls the corresponding control valve, is in communication with the corresponding channel, so that the purpose of automatic sorting and conveying of the bags is achieved, the labor force is liberated, and the productionefficiency is improved.

Sorting device for classifying bags

The invention discloses an intelligent sorting device based on machine vision and a sorting method thereof. The device comprises sorting transmission mechanisms, manipulator sorting mechanisms, support frames, visual sorting mechanisms and a PLC (programmable logic controller), wherein each sorting transmission mechanism comprises a feeding station, a horizontal flow detection station, a transportation station, a transfer discharging station and a horizontal flow station; each visual sorting mechanism comprises a packaging state system, a shape discrimination system, a color discrimination system and cameras; each packaging state system, each shape discrimination system and each color discrimination system share a station of each camera; each camera is arranged on the outer wall of the bottom of each support frame; packaging state of an article is checked and classified by each packaging state system, a target article is locked and shunted by combining with the color and the shape; andthe main functions of the cameras are positioning, identifying and controlling the manipulator sorting mechanism to work through the PLC. According to the device, the problems of complex manual operation, low efficiency and difficulty in sorting of the prior logistics parcel sorting art are solved.

Intelligent sorting device based on machine vision and sorting method thereof

The invention discloses a crop seed quality detection and screening system. The crop seed quality detection and screening system comprises a support, a material bearing conveyer belt, a funnel, a mechanical grabber, an image acquisition component and a computer, wherein the support comprises two cross rods and an inverted-U-shaped support; the material bearing conveyer belt is fixed on the cross rod at the bottom of the support; a material containing disc is arranged on the end of the material bearing conveyer belt; the funnel is fixedly arranged on the cross rod at the top of the support, andthe material leaking speed of the funnel is controlled through a material leaking control motor which is fixedly arranged on the cross rod on the top; a material outlet in the funnel is arranged on the upper part of the material bearing conveyer belt; the mechanical grabber is fixedly arranged on a vertical rod of the support, and is controlled by the computer to array and screen seeds on the material bearing conveyer belt; the image acquisition component is used for acquiring images of the seeds on the material bearing conveyer belt; and the computer performs seed identification and classification according to received seeds images according to a seed identification model which is built based on a convolution neural network, and outputs control signals to the material leaking control motor and the mechanical grabber. By adopting the system, seed sorting efficiency and accuracy are improved.

Crop seed quality detection and screening system

The invention discloses logistics sorting equipment with an anti-jolting structure. The logistics sorting equipment comprises an A conveying belt module, a C-shaped support, an inclined table, a C conveying belt module, a base, a B conveying belt module and a control module. The A conveying belt module and the B conveying belt module are close to each other and are flushed. The C conveying belt module is located below one side of the B conveying belt module. The inclined table is mounted between the B conveying belt module and the C conveying belt module. According to the equipment, through the arrangement of an angle adjusting mechanism, when the direction of bar codes on the two sides of a to-be-sorted object located on the B conveying belt module is opposite to the motion direction of aconveying belt, the angle adjusting mechanism works, a friction pressure plate is in contact with the upper surface of the to-be-sorted object, the friction pressure plate and the to-be-sorted objectrotate together by a fixed angle, a code scanner A can normally acquire information of the bar codes on the to-be-sorted object, then the angles of each to-be-sorted object on the B conveying belt module are the same, and subsequent sorting is convenient.

Logistics sorting equipment with anti-jolting structure

The invention discloses a fruit quality detection system and method based on depth learning. The fruit quality detection system and method based on depth learning mainly judge the fruit quality according to the color, texture, shape and size of the fruit surface to complete the classification. The whole detection system is composed of a fruit transmission system, a computer vision recognition system and a classification system. Compared with a traditional fruit detection system, the system has the distinct characteristics of deep learning, and the characteristics are mainly reflected in the fruit image processing. Through the detailed analysis of a network structure of a convolution neural network algorithm and the training process of the convolution neural network, a fruit image recognition system based on the convolution neural network is constructed. Compared with a general neural network algorithm, the fruit image recognition system has the characteristics of simple structure, lesstraining parameters, high adaptability and the like.

Ma Zichen

Papers

Fruit quality detection system and method based on depth learning