The goal of this thesis is to make a portable system for instrument calibration which is partially automatized. The system consists of a web camera connected to a myRIO device which contains a program written in LabVIEW. The main components of the program are the virtual instruments of Vision and motion library: Vision Acquisition Express VI for acquiring the device's screen image, Vision Assistant Express VI for editing the image and preparing it for digit recognition using IMAQ Read LCD VI. The first step is to establish a connection between the main program and the Izvor_AX program using Network Stream functions and setting the source's reference value to be read. After a few moments, devices we are calibrating measure this value and then Vision Acquisition Express VI forwards the camera image of devices under test to the monitor. The image is converted to black and white format to be suitable for digit recognition process. In the following step the user is required to define regions of interest on the image, using the subroutine Inicijalizacija, which also defines the number of devices we intend to calibrate. Based on the defined regions of interest, virtual instruments analyze the data. In the final step the digit recognition process is performed. The subroutine Sto_pisati defines the sign value while the decimal point position is being determined based on the user input using the subroutine Decimala. After that, the system takes readings in real time. By turning the switch „Kalibracija“, the percentage error of an individual device is calculated and stored in the MjerniPodatci.txt file on the myRIO device. The stored data can be accessed via internet for analysis in Excel