JPH01259243A - Method and instrument for automatic determination of microscopic image of opaque ore or the like - Google Patents

Abstract

PURPOSE:To execute the discrimination and identification of opaque ore and to execute easy and rapid ore identification and binarization by numerically converting a microscopic image and RGB images to IHP data, and identifying and binarizing the ore by the upper limit value and lower limit value of each of IHP 3 elements. CONSTITUTION:The texture image of the polished slice sample of a material contg. the opaque ore to be analyzed is observed by a microscope 11 and the texture image by the microscope 11 is picked up by a color TV 12. The picked up image data is numerically converted by an AD converter 13 to ternary systems; lightness (I), hue (H) and saturation (P). The converted information is stored in a memory 14. The identification and binarization of the ore by the upper limit value and lower limit value of each of the IHP 3 elements of the RGB images picked up by the camera are executed by a central control processing unit 15 in accordance with such information and, therefore, the discrimination and identification of the opaque ore are executed. In addition, the IHP is stored, by which the identification and binarization of the ore are easily and rapidly executed by utilizing the values thereof at every movement of the visual field.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method and apparatus for tissue analysis by image analysis of substances containing opaque minerals, and in particular, a quantitative measurement method for automatically measuring the phosphoric acid ratio of each tissue. and regarding measuring devices.

[Prior art]

Metal ores usually contain useful metals such as sulfide minerals and oxide minerals, as well as unnecessary non-metal oxides such as non-metal oxide minerals, and the composition ratio of these mineral structures and cross-sectional particle size distribution vary. Quantitative measurement results are a fundamental factor in evaluating the properties of ores and determining various conditions such as crushing and separation during ore processing.

Conventionally, the metal sulfide minerals or metal oxides described above are optically opaque in the form of thin pieces for measurement, so the thin pieces are solidified with resin, the surface is mirror-finished, and the color tone is determined under a reflective microscope. Tissues are identified by manual point counting, line analysis, or area analysis.

In these methods, the main operation is performed by the inspector, which involves selecting multiple fields of view of tissues representative of the target sample, identifying each mineral structure, and determining the area ratio of the tissues (area analysis method).
, record, repeat these steps preferably for 50 or more visual fields, calculate the integrated value, and multiply the area ratio (volume ratio) by the specific gravity of each mineral. The weight ratio of the tissue is being determined.

Since the above conventional method is basically based on visual observation,
Inevitably, large variations in measurement accuracy, measurement speed, etc. due to purity, degree of fatigue, individual differences, etc. in the work of inspectors were unavoidable.

As a means to deal with this, an automatic measuring device has been developed that is used to automate measurements regarding reflection microscopic structures. For example, a reflection microscope image is taken with a color 1mm camera, the examiner sets a threshold, the desired image is binarized using this threshold, and a binarized image of the desired tissue is extracted. However, methods are being used to find the area using an appropriate method.

[Problem to be solved by the invention]

The automated method for measuring reflection microscopic tissues described in the conventional technology binarizes each tissue using only black and white and gray thresholds, so two types of tissue with different hues but equal brightness (light reflectance) are generated. The basic problem was that it was not possible to distinguish between three different minerals, and that it was not possible to reproduce the visual judgments made by inspectors, including discrimination based on hue.

The present invention solves the above-mentioned problems and uses a device to reproduce with high precision the conventional judgment made by inspectors that makes full use of all three color elements of hue, brightness, and saturation.
Moreover, it is an object of the present invention to provide a method and apparatus for automatic quantitative measurement of microscopic images of opaque minerals, etc., which can be easily automated.

[Means to solve the problem]

The method for automatic quantification of microscopic images of minerals according to the present invention which achieves the above object is to observe the U-textured scratches of a polished sample of a material containing opaque minerals under a microscope, and to color the microscopic images of the microscopic images. The camera captures images, red (R), green (G).
), blue (B) three-color spectrum hand image data is numerically converted into a three-component system of brightness (■), hue (H), and saturation (P), and the upper and lower limits are calculated for each IHP element. The upper and lower limits are set to the extent that the desired mineral structure can be identified for the range limited by two thresholds or vice versa, and each binary IHP value is The product set of one or more of the images is converted into a binarized image by IHP, and a binarized image by IHP obtained for several different minerals is obtained, thereby determining the factors of the target mineral in the field of view. is calculated and measured, and the upper and lower limits of IHP for each target mineral are stored, and subsequently, for another field of view, the above-mentioned imaging, IHP exchange, and IHP based on the stored upper and lower limits are performed. The method includes repeating the procedure involving binarization and measurement of the factors and integrating the measurement results to measure the desired material for the entire sample.

Furthermore, the apparatus according to the invention for carrying out the above method comprises:
A microscope for observing the tissue image of a polished piece sample of a material containing an opaque mineral whose structure is to be analyzed, a color TV for capturing the tissue image using the microscope, and a color TV for capturing the image data from the color TV in terms of brightness (I) and hue ( H) respective conversion devices for numerically converting into a three-component system of chroma (P), a storage device for storing information from the conversion device, a central control processing device for processing the information, and the central control processing. It includes an image display and storage device that receives information from the device, and an automatic stage focus control device.

[Effect]

Since it is configured as above, the IH of the RBG image captured by the TV camera is
P Since minerals are identified and binarized using the upper and lower limit values for each of the three elements, it is possible to distinguish and identify opaque minerals, and by recording the upper and lower limit values of IHP, the field of view can be improved. Each time the value is moved, mineral identification and binarization can be repeated easily and quickly, and various desired data such as the area ratio of each mineral can be measured using any known software.

Display and printing can be performed easily and quickly, and these steps can be automated if necessary.

〔Example〕

An embodiment of the apparatus for carrying out the method for analyzing a tissue containing opaque minerals according to the present invention will be described below with reference to the drawings.

FIG. 1 is a perspective view of a polishing piece used for analyzing the structure of ore. The polished piece 1 is made by solidifying a representative part 2 taken from an ore sample to be measured by a conventional reduction method and solidifying it with a synthetic resin 3, and then polishing the sample surface to form a measurement surface IA. . In the present invention, such a polishing piece 1 is used to measure the composition ratio of the tissue constituting the sample to be measured.

FIG. 2 shows a schematic diagram of the apparatus of the present invention.

In FIG. 2, a microscope 11 measures the structure of a polished piece l of a target mineral, a color TV 12 takes a mm image observed with the microscope, and image data from the color TV 2 is displayed in terms of brightness (I), two hues (H ), chroma (P), a memory device 14 for storing information from the AD converter, a central control processing unit 15 for processing the information, and a central processing unit 15 for processing the information. Color image display device 17 receiving information from control processing device 15
, a graphics terminal including a graphics link 18 and a keyboard, and an automatic stage focus controller 22 for controlling the stage device 10 disposed on the microscope from the central control processing unit.

Next, regarding the tissue analysis method according to the present invention, the apparatus having the above configuration will be described as follows.

In FIG. 3, which is a flowchart showing an outline of the method of the present invention, first, any mineral containing as many mineral species as possible is selected on the surface to be measured of the polishing piece 1, and observed on a CRT (optional). ) and set the initial conditions including the objective lens with magnification suitable for measurement, illumination intensity, and illumination filter...3
1. Red (RED) and green (
32. This memory is stored in the main unit, and is expressed as a real color image in the cathode ray tube (CRT) 17 with each intensity of RGB. . This data is calculated and converted into IHP combination data, and... 33.Next, select only the minerals to be measured while comparing the real color image and the binarized image on the CRT.・ ・34゜ Next, set the upper and lower limit levels of IHP that can be binarized to a sufficient degree for that mineral... 35゜At this time, specify any pixel on the target mineral on the CRT 17 and select the corresponding By using the method of displaying each value of IHP of a pixel, each upper and lower limit value of IHP can be determined more easily and quickly.

Memorize the upper and lower limits of IHP for identifying the first target mineral...End.

After that, repeat the same operation for the second target mineral, and
Determine and record the upper and lower limits of HP.

The same operation is performed for the third, fourth, etc. necessary types of target minerals to identify them, and the binarization data is stored. If all necessary mineral identification data (data for binarization) cannot be obtained with only the first field of view, select an appropriate second field of view to compensate for this, and select RGB? r
Repeat the process of importing the A light image, converting it to IHP data, and determining and storing the data for mineral identification binarization. This completes the preparation for continuous measurement.

Next, the image of the field of view that has already been captured and converted to IHP data is binarized using the mineral identification binarization data stored in the above preparation stage to obtain a binarized image for each mineral. . This binarized image is usually stored in a binarized image memory.
(hereinafter referred to as BM) is stored for each mineral, and parameters such as area, maximum length, circular phase diameter, etc. of each partial image (particle) in that BM are measured, particle size distribution, area ratio, volume and weight. Calculate the conversion data. Estimated quality can also be determined by roughly inputting the specific gravity and component analysis values of each mineral.

Subsequently, another arbitrary field of view of the polished piece is selected by the automatic (or manual) stage device 22... 31
, import as a grayscale image using RGB... 32,
Convert to IHP data... 33. Obtain a binarized image easily and quickly using the mineral identification binarized data stored above... 37. Area ratio, weight ratio, estimation Measure quality and particle size distribution, and print if required. ... 38. By repeating this operation of changing the field of view a predetermined number of times (usually 50 times), storing and calculating the data in an integrated manner, the area ratio of each mineral in the entire measurement sample,
The weight ratio, estimated grade, and particle size distribution of the mineral cross section can be determined and displayed. Each time the field of view moves, the focus is adjusted by an automatic (or manual) focusing device 22 as necessary.

〔Effect of the invention〕

As explained above, the present invention is applicable to microscopic images, color TV
By numerically converting the RGB image taken by a camera into IHP system data, and identifying and binarizing minerals using the upper and lower limit values for each of the three IHP elements, it is possible to distinguish and identify opaque minerals, which was difficult with conventional methods. In addition, the upper and lower limits of IHP can be stored in a storage device, and these values can be used each time the field of view is moved to easily and quickly repeat mineral identification and binarization. If possible, using any known software, the area ratio, weight ratio, estimated quality of the sample, particle size distribution of the mineral cross section, etc. of each mineral can be easily measured, displayed, and printed.
An extremely large effect can be obtained in that it can be quickly and automatically performed as needed.

[Brief explanation of the drawing]

FIG. 1 is a perspective view showing a polishing piece used in quantitative measurement of tissue according to the present invention. FIG. 2 is a schematic configuration diagram showing the configuration of an automatic quantitative measurement device for mineral structure according to the present invention. FIG. 3 is a flow diagram outlining the method of the present invention. 1... Polished piece 2... Ore particles 3...
・Resin IA・ Measurement surface 10... Stage device 11... Reflection microscope 12... Color TV
13...AD converter 14...Memory
15... Central control processing unit 16... Ink face 17... CRT 18... Graphic link 19... Graphic terminal 20... Keyboard 21... Mouse 22 ・
Auto stage autofocus controller 31... Initial conditions (objective lens magnification, illumination intensity 1 illumination filter) settings 32, RGB gradation image capture 33, RGB gradation image I HP conversion 34... Target mineral selection 35... Binarization level (upper and lower limits) setting 36...Measurement item setting 37... Binarization 38...Measurement/display...
Printing 39... Movement of field of view 40... Terminated patent applicant Sumitomo Metal Mining Co., Ltd. Nireco Co., Ltd.

Claims (1)

[Scope of Claims] 1. Observe the tissue image of a polished piece sample of a substance containing opaque minerals using a microscope, and capture the tissue image using a color TV camera. ), blue (B)
Image data based on the three color spectrum bands of brightness (I)
, numerically converted into a three-component system of hue (H) and saturation (P), and I
For each HP element, set the upper and lower limits to the extent that the desired mineral structure can be identified, and set the range limited by two thresholds, the upper limit and the lower limit, or vice versa. 1 of each IHP binarized image binarized by value
The product set of three or more images is converted into a binarized image by IHP, and the IHP binarized images obtained for several different minerals are obtained, and the factor of the target mineral in the field of view is calculated and Measure, memorize the upper and lower limits of IHP for each target mineral, and then continue to measure the above for another field of view.
repeating the procedure including imaging, IHP exchange, IHP binarization with stored upper and lower limit values, and measurement of the factors, and integrating the measurement results to measure the desired material of the entire sample; A method for automatic quantitative measurement of microscopic images of minerals. 2. A microscope for observing a tissue image using a polished sample of a substance containing an opaque mineral whose structure is to be analyzed; a color TV for capturing the tissue image using the microscope; each conversion device for numerically converting into a three-component system of hue (H) and saturation (P); a storage device for storing information from the conversion device; a central control processing device for processing the information; 1. An automatic quantitative measuring device for microscopic images of minerals, comprising an image display and storage device that receives information from a control processing device, and an automatic stage/focus control device.