SU1737470A1 - Device for selecting images of point objects - Google Patents

Description

The invention relates to automation and computing and can be used in pattern recognition systems.

The purpose of the invention is to improve the accuracy of the device.

FIG. 1 is a block diagram of the device; in fig. 2 - discriminator circuit; in fig. 3 shows the arrangement of the image elements in the aperture formation unit.

The device for selecting images of point-like objects (Fig. 1) contains image sensor 1 connected in series, analog-digital converter 2, digital median filter 3, aperture shaping unit 5, first 6 and second 7 discriminators, OR 8 element, analog adder 9.

The discriminator (Fig. 2) contains an adder 10, a group of comparators 11 and the element 12.

The device works as follows.

The video signal during progressive scanning by the image sensor 1 is fed to the input of analog-digital converter 2. The clock frequency of converter 2 is chosen so that the distance between two samples is half the length of the image line of an ideal point object. At the same time, a real point object is characterized by two or three samples per line, and the size of its two-dimensional image does not exceed three elements.

Digital median filter 3 removes one-element pulse noise per line that can be mistaken for images of point objects.

The aperture shaping unit 5 generates signals from image elements that fall into an aperture of size 5x5 (Fig. 3). Signals from the first output (x0) and second

cl

with

Xi

with

Xi

;four

Xi

ABOUT

outputs (x1 ... x1b) unit 5 simultaneously arrive at the inputs of the first 6 and second 7 discriminators.

Selection of images of point objects is carried out using discriminators 6, 7 (Fig. 2) according to the following algorithm. In summer 10, the magnitude of the central element of the aperture X0 is added to the threshold value P, which is predetermined by calculation. In the first 6 discriminator, this value is negative Pi P, in the second 7 discriminator it is positive P2 P. The output of the adder 10 is compared on the comparator 11 with the values of the extreme elements of the aperture xi ... xie. In this log. 1 at the outputs of the comparators 11 are produced

under the conditions x0 - Р xi, i 1

16 for the first 6 discriminators and x0 + P xi,

i 116, for the second 7 discriminator.

Simultaneous fulfillment of the conditions by all comparators means that the image size of the object does not exceed the size of the central part of the aperture shaded in FIG. 3, i.e. three elements. In this case, the output of the discriminator 6 or 7 will be installed log. 1. Thus, in accordance with this algorithm, the first 6 discriminator generates a signal with a log level. 1 in the case when its inputs receive signals from a point object, the brightness of which is greater than the brightness of the local neighborhood (positive contrast), and the second 7 discriminator generates a log. 1, when the brightness of a point object is less than the brightness of a local neighborhood (negative contrast).

The signals from the outputs of the first 6 and second 7 discriminators through the element OR 8 are fed to the input of the analog adder 9, to another input of which the video signal from the image sensor 1 is fed. In this case, at the output of the analog adder 9, the signal will correspond to the original image, in which near point objects there will be labels of constant luminance.

Claims (1)

An apparatus for selecting images of point objects comprising serially connected image sensors, an analog-digital converter, a digital median filter, a first discriminator and a synchronizer, the output of which is connected to the synchronizing inputs of the analog-digital converter and the digital median filter, characterized in that in order to improve the accuracy of the device, a second discriminator, an OR element, an analog adder, an aperture shaping unit, whose input The output of the digital median filter and the outputs are connected to the information inputs of the first and second discriminators, the outputs of which are connected to the inputs of the OR element, the output of which is connected to the first input of an analog adder, the second input of which is connected to the output of the image sensor, and the output is the output of the device, The adder overflow input is the device control input.