JPS5952775B2 - flaw detection equipment - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a flaw detection device that photographs the surface of a hot steel material using an imaging device, and more particularly to a flaw detection device that detects flaws in the horizontal scanning direction of the imaging device.

In steelmaking plants, defect detection during the blooming or continuous casting process is generally carried out by cooling the hot steel material (hereinafter referred to as hot steel material) to room temperature and then clearly observing it in the cooled state. It is.

For this reason, it is necessary to heat the steel material again after the flaw detection step, which is one of the factors that prevent the cost of steel materials from being reduced. In order to solve this problem, it would be sufficient to detect flaws in the hot state, but this can be done simply by photographing the surface of the hot steel material using an imaging device and extracting flaw signals directly from the video signal. Automated equipment that does this cannot identify defects with sufficient accuracy.

That is, when a hot steel surface is photographed, the video signal from the imaging device becomes a waveform with an amplitude corresponding to the brightness of the steel surface.
Since a flawed portion generally has higher brightness than a non-flawed portion, the amplitude tends to be larger than that of a flawed portion. However, it is often difficult to distinguish between irregular noise and defect signal components, which are often included in a video signal, based only on their amplitude levels. SUMMARY OF THE INVENTION An object of the present invention is to provide a flaw detection device that is capable of reliably detecting flaws on the surface of a hot steel material by distinguishing between flaw signal components and irregular noise in the horizontal scanning direction of an imaging device.

When there is a flaw 2 on the surface of the hot steel material 1 as shown in FIG. 1a,
When this is horizontally scanned in the direction of arrow A by an imaging device, the video signal is a portion A corresponding to flaw 2, as shown in figure b.
appears as a waveform with a larger amplitude than in part B corresponding to the surface without flaws.

Therefore, waveforms A and B can be distinguished in terms of amplitude levels. However, if the video signal contains irregular noise C, it is difficult to distinguish between the noise C and the defect signal component A simply based on the amplitude level. However, the noise C generally has a shorter time width than the flaw signal component A. Therefore, in the present invention, it is determined whether the waveform is a flaw signal component or irregular noise based on the time width of the waveform having a predetermined reference amplitude level D or higher in the video signal. Note that E is a horizontal synchronization signal. At this time, time width detection is performed digitally.

That is, the video signal is binarized with a reference amplitude level D, and
The converted signal is sampled and converted into a code string of logic "1" (above the reference amplitude level D) and logic "0". Then, it is determined whether it is a defect signal component or irregular noise based on the number of logic bits of "1" in a continuous predetermined number of bits. minutes, where n is a number based on the Someya-Shannon sampling theorem.In addition, if the sampling period is selected appropriately, there is a high possibility that irregular noise will be located between the sampling points, and It is also possible to remove irregularity noise at a stage before width determination.An embodiment of the present invention will be described below based on the plot diagram of FIG.

In the figure, reference numeral 11 denotes a buffer amplifier that amplifies the power of a composite video signal a from an imaging device photographing the surface of a hot steel material, and its output is led to an analog comparator]2, which outputs an analog amplitude and set value b. (Equivalent to the reference amplitude level D)
compared to The output of analog comparator 12 is logic “1”
It is a binary signal in which the section (video signal section exceeding the reference amplitude level b) and the logic "0" section are continuous, and the AND circuit 1
3 is supplied to one input. Logical AND. A sampling pulse from a timing generation circuit 14 is led to the other input of the circuit 13, and the output of the analog comparator 12 is sampled by the sampling pulse. The timing generation circuit 14 generates a composite video signal using a synchronization separation circuit 15. Based on the phases of the horizontal and vertical synchronization signals extracted from a, the output of the reference oscillator 16 is used to generate various timing signals including the sampling pulse. Thus, the output of the AND circuit 13 is serially transferred to the shift register 17 in synchronization with the sampling pulse.

Reference numeral 18 denotes an adder that adds (counts) the number of logical ゜゜1゛ in a certain consecutive determined number of bits (hereinafter referred to as N bits) in the shift register 17; are supplied in parallel.

Therefore, shift register 17 requires at least N bits, but if the number of bits is greater than this, only its N output bits are supplied to adder 18. 19 is adder 1
This is a digital comparator that compares the output of adder 18 with a digital set value C, and outputs flaw information output d when the output of adder 18 within one horizontal scanning period is greater than the set value C.

The flaw detection apparatus having the above configuration has the advantage that the video signal a can be digitized in real time by the analog comparator 12 and the AND circuit 13, and a high-speed A/D converter is not required.

Furthermore, if the output of the adder 18 does not reach the digital setting value C, even if there is logic ゜゜1゛ in the N bits, the defect information signal d is not sent out, so that equivalently the irregularity noise is removed. Therefore, according to the present invention, it is possible to reliably detect flaws on the surface of a steel material in a hot state, thereby eliminating the step of cooling and then heating for flaw detection, thereby reducing costs.

It should be noted that the present invention is not limited to the above embodiments, and can be modified in various ways.

[Brief explanation of the drawing]

Fig. 1a is a partial plan view showing the steel surface in a hot state, Fig. 1b is a composite video signal waveform diagram for one horizontal scanning period photographing the steel surface in Fig. FIG. 3 is a block diagram showing an example. 11... Buffer amplifier, 12... Analog amplitude comparator, 13... AND circuit, 14... Timing generation circuit, 15... Synchronization separation circuit, 16... Reference oscillator, 17...・Shift register, 18... Adder, 1
9...Digital comparator.

Claims (1)

[Claims] 1 A composite video signal obtained by photographing the surface of a hot steel material is compared with a predetermined reference level and binarized, and this signal is sampled at a predetermined repetition period and sequentially transferred to a shift register. and means for determining the number of bits of one logic label included in a plurality of consecutive binary signals within one horizontal scanning line based on the stored binary signals, and determining flaws on the surface of the heated steel material. A flaw detection device that is characterized by: