JPH06265408A - Method for making spectral diffraction of image and its device - Google Patents

Abstract

PURPOSE:To precisely acquire a spectral image with optical conditions kept unchangeable as far as possible by relatively moving a continuous interference filter between a measuring object and a two-dimensional image sensor for detecting an image. CONSTITUTION:The movement of a continuous inteference filter 6 is executed by a DC motor 32 so as to repeat a movement for the width of one specified wave length zone and stoppage required for charging a CCD (solid image pick up element) sensor 7 by a transmitted light. A timing for starting the above actuation is given by detecting the introduction of an apple 4 onto a measuring stage through a detector. The sensor 7 is made to receive the light through the filter 6, and an electric charge charged in each picture element in proportion to the quantity of received light turns into image data, and is read out synchronously with the moving timing of the filter 6 at fixed timing from the sensor 7 into the picture element, and information required for spectral image processing is thereby sent from a control section 8 to a central processing circuit 36. Thus image pick-up can be carried out without the variation of optical conditions to obtain highly precise spectral image.

Description

Detailed Description of the Invention

[0001]

The present invention relates to reflected light or object transmitted light obtained from an object represented by various solid materials such as plants such as agricultural products, animals such as fish, minerals and industrial products. The invention relates to a method and a device for obtaining a spectral image of the object based on it.

[0002]

2. Description of the Related Art Recently, an optical technique, which is one of the methods for measuring and analyzing the physical properties and properties of an object, is rapidly advancing. For example, as a technique for analyzing attributes such as physical properties and properties of an object based on imaging information, a method of spectroscopically analyzing a minute portion of an object to be measured has been known for a relatively long time. A method called image spectroscopy (imaging spectroscopy), which combines image optical technology and spectral analysis technology, is proposed in which a large number of spectrally captured images, that is, spectral images are obtained by using a CCD (solid-state imaging device) or the like. ing.

The former spectroscopic analysis method for a minute portion has been well known as a method for obtaining analysis information by dispersing a minute portion into each wavelength range, and a method devised for spectral analysis, for example, a continuous wavelength tunable filter is used. Japanese Patent Application Laid-Open No. 57-1518
The method of No. 30 is also proposed.

On the other hand, the latter image spectroscopic method is still in the research stage, and there are few proposals that can be put to practical use as it is, but in general, a combination of the spectroscopic method and the image optical technique is used. This is a method to capture a large number of spectroscopic images by repeatedly imaging the entire measurement target (or a wide area of a part of the object), and by comparing the obtained spectroscopic images, only the spectroscopic analysis of the minute parts described above is performed. Since it becomes possible to obtain useful information that cannot be obtained by the above method (for example, the sugar content distribution of the whole fruits and vegetables in the study of the present inventors), much attention has been paid recently.

Explaining the usefulness of this image spectroscopy in more specific cases, taking fruits and vegetables such as vegetables and fruits such as peaches and apples, the production-collection-selection-shipment-sales of In the process, even if the varieties are the same, the size and shape of each individual may differ depending on the environmental conditions, the presence or absence of scratches, and the internal quality such as sweetness of peaches and apples may vary. It is well known that the appearance quality, internal quality, etc. are directly reflected in the transaction price in the market because they are the judgment factors when the consumer purchases them. Is. Therefore, if information on these qualities, especially internal qualities that were conventionally difficult to measure and inspect, is obtained, it will be an advantage that consumers can purchase fruits and vegetables as expected, and producers will be able to meet their production efforts. It can be said that it is desirable in the industry because it will earn more income and will increase the motivation to produce higher quality fruits and vegetables.

However, since the image spectroscopy method, which is effective as a method for evaluating the internal quality, is still in the research stage as described above, even if it can be applied to an extremely limited field, For example, it is not suitable to carry out the selection of fruits and vegetables on an industrial level.

[0007] For example, one of the image spectroscopic methods proposed so far is "two-dimensional spectroscopic analysis method using digital image".
(Opttronics No. 2 (1992)) is proposed, but as shown in FIG. 9, a plurality of single colors (single color) having different transmission wavelengths are provided between the two-dimensional image sensor camera 200 and the fruits and vegetables 201. Wavelength band) Interference filter 2
02 can be sequentially arranged by rotation of the disk 203, and a large number of spectral images can be obtained by sequentially capturing images for each of the switched monochromatic interference filters 202. For the purpose of
Since the number of monochromatic interference filters for spectroscopy increases, it is inevitable that the switching mechanism becomes large and the mechanism becomes complicated, and since the filters 202 are sequentially imaged while switching, the imaging time for one individual becomes long. Therefore, in the actual fruit-picking work in which a large number of fruits and vegetables are transferred for example on a conveyor (or the fruits and vegetables are temporarily stopped on the measuring stage) and image pickup is performed for quality measurement, the efficiency of the fruit-picking work is remarkable. As a result, it is not practically applicable.

[0008]

SUMMARY OF THE INVENTION As described above, in the image spectroscopic method which is carried out by the method of sequentially switching a large number of monochromatic interference filters, an industrial scale (for example, selection) is employed in view of speed and work efficiency. It is unsuitable for implementation in (work).

The present invention solves the above problems and provides a novel image spectroscopic method capable of rapidly capturing a large number of spectroscopic images from which excellent evaluation information can be obtained in the quality evaluation of fruits and vegetables. Another object of the present invention is to provide a simple and compact image spectroscopic device that can be used for implementing this method.

Another object of the present invention is to provide a novel spectroscopic image method capable of accurately obtaining a spectroscopic image which is a basis for quality evaluation without changing the optical conditions for obtaining the spectroscopic image as much as possible. There is a device to be used.

Still another object of the present invention is to apply the method and apparatus realizing the above object to a fruit selection technique for quality evaluation of fruits and vegetables, whereby a large number of objects are continuously transferred on a conveyor or the like. It is an object of the present invention to provide a method and an apparatus for evaluating the quality of fruits and vegetables, which enables rapid and high-efficiency evaluation of fruits and vegetables and enables selection on an industrial scale.

[0012]

The present inventor has completed the present invention described in each of the claims in order to achieve the above object, and one of the features is particularly , A two-dimensional continuous interference filter (also called a continuous wavelength filter or a continuous wavelength tunable filter: hereinafter simply referred to as "continuous interference filter"). This continuous interference filter has a narrow area provided so as to transmit only light of a specific wavelength (having a certain wavelength band width), for example, in one direction of a plane (flat plate) having a rectangular two-dimensional spread,
Although the divisions are divided and connected, and the signal processing procedure is not particularly limited, the transmission wavelengths specified for each of these connected areas are continuous or stepwise. It is provided to change to.

Further, according to the present invention, the continuous interference filter is moved in one direction between the two-dimensional image sensor for detecting the light transmitted through the continuous interference filter and the object to be measured, so that each of the two-dimensional image sensors is moved. Another feature is that the pixels are made to sequentially detect the filter transmitted light whose transmission wavelength changes.

In the image spectroscopic method according to claim 1 of the present invention having these characteristics, a narrow region which transmits only light of a specific wavelength is connected by dividing a section in one direction of a surface having a two-dimensional spread. A plurality of continuous interference filters provided so that the transmission wavelengths of the light respectively specified for each of the concatenated narrow regions are continuously or stepwise changed along the one direction. And the two-dimensional image sensor are moved relative to each other in the above-mentioned one direction, whereby each pixel of the two-dimensional image sensor causes the light from the measurement object to be transmitted at each wavelength as filter transmission light whose transmission wavelength changes. It is characterized in that they are sequentially detected.

According to the second aspect of the invention, the signals corresponding to the common transmission wavelength of the transmitted light detected by each pixel are integrated to create a spectral image for each wavelength. It is characterized by having done.

It goes without saying that the light from the object to be measured may be reflected light or object transmitted light.

Further, in the image spectroscopic device according to the fourth aspect of the present invention, a large number of narrow regions, which pass only light of a specific wavelength, are divided and connected in one direction on a surface having a two-dimensional spread. , The transmission wavelengths of the light respectively specified for each of the connected narrow regions are provided so as to change continuously or stepwise along the one direction, and one surface is opposed to the measurement object. A continuous interference filter, a two-dimensional image sensor arranged so as to face the surface on the opposite side of the continuous interference filter, and the continuous interference filter is relatively moved in the one direction with respect to the measurement object and the two-dimensional image sensor. After photoelectrically converting the filter transmitted light which is transmitted through the filter moving means and the moving continuous interference filter and is received by each pixel of the two-dimensional image sensor, it is sequentially read from the pixel as a signal. Characterized in that a to signal readout means, in addition, by integrating a signal corresponding to the common transmission wavelength from among the read-out signals from the pixels,
Another feature is that it is configured to include an image processing unit that creates a spectral image for each transmission wavelength.

The continuous interference filter used in the present invention is formed by film-forming the dielectric film so that the film thickness of the dielectric film gradually increases in one direction in which the continuous interference filter moves, or by dividing a partition into one direction in a rectangular flat plate. It is possible to exemplify a structure in which narrow regions are connected to each other, and a monochromatic interference filter having a different transmission wavelength is provided in each region as an integrated body. More preferably, the transmission wavelength is changed stepwise by providing monochromatic interference filters having the same transmission wavelength with gradually different inclination angles from the posture orthogonal to the light transmission direction (hereinafter referred to as the inclination type). Filter) can be used,
Further, a plurality of monochromatic interference filters having different transmission wavelengths and this inclined type configuration may be used together. The inclined continuous interference filter can be easily manufactured by gradually inclining the posture of the monochromatic interference filter and fixing it with a transparent body. According to this tilt type, a monochromatic interference filter of a single standard or a plurality of standards with different specific wavelengths is used, and if necessary, in increments of several nm to several 100 nm.
It is possible to manufacture a continuous interference filter in which the transmission wavelength is gradually changed over a range of 00 nm to several 1000 nm or more.

The transmission wavelength range of the light transmitted by the continuous interference filter is not particularly limited, but for the purpose of quality inspection of fruits and vegetables, the visible light range to the near infrared range (550 to 550). In many cases, it is appropriate to set it to about 1100 nm).

Examples of the two-dimensional image sensor include a CCD image sensor and the like.

In the practice of the present invention, the structure in which the object to be measured and the two-dimensional image sensor are fixed and the continuous interference filter is moved with respect to these is particularly preferable and recommended because of its simple structure. For example, in the case where fruits and vegetables are transferred on a conveyer to perform selection, the above requirement can be met by providing a measurement stage for temporarily stopping the fruits and vegetables in the middle of the conveyer path to perform imaging. . To stop the movement of fruits and vegetables, the conveyor may be stopped intermittently or the tray may be stopped by a stopper. In addition, if you do not want to stop the fruits and vegetables once, move the CCD camera consisting of a two-dimensional image sensor (and if necessary, the illumination optical system) in synchronization with the conveyor, and The continuous interference filter may be moved relative to the two-dimensional image sensor.

The movement of the continuous interference filter is not particularly limited, but an intermittent movement combining a movement of a narrow wavelength region divided by a servo motor, a pulse motor or the like and a stop for imaging is used. It is preferable to adopt the method, and it is preferable to synchronize the reading of the signal photoelectrically converted by each pixel of the two-dimensional image sensor.

According to the present invention having the above-described structure, the optical condition between the measuring object and the image of the measuring object projected on the two-dimensional image sensor which is the image sensor is always constant. With the structure maintained, the continuous interference filter is moved so that it traverses the image projection optical path, and according to the structure in which the transmission wavelength range of the continuous interference filter changes, the spectrum of each pixel of the two-dimensional image sensor is received in time series. Will be done. It should be noted that although a similar spectral image can be obtained by moving the two-dimensional image sensor while the continuous interference filter and the measurement object are fixed, the measurement object and the two-dimensional image sensor move relatively. Needless to say, the configuration of the present invention is preferable in consideration of the problem that the optical conditions fluctuate as a result.

A two-dimensional description will now be made with reference to FIGS. 1 and 2 which schematically show the relationship between the fundamental structure required for the image spectroscopy performed by the present invention and the spectral image obtained thereby. In the image sensor 101, the moving direction of the continuous interference filter is set to be vertical, and the direction orthogonal to this is set to be horizontal, and the two-dimensional image sensor 101 is set to 1, 2, 3 ... i, i + 1 as shown in FIG. , ... mth (vertical), 1, 2, 3 ... j, j + 1, ... nth (horizontal) numbered pixel matrix, and another element, continuous interference Filter 102
However, the transmission wavelength range divided into the narrow region is gradually changed from the wavelength λ ₁ (for example, a wavelength range of 550 ± 10 nm) at the tip in the moving direction toward the long wavelength side by Δλ, and Assuming that the wavelength is divided stepwise up to a wavelength λ ₂ (for example, a wavelength range of 1100 ± 10 nm), the (i, j) th pixel of the two-dimensional image sensor 101 (see FIG. 1).
Λ ₁ to λ as the continuous interference filter 102 moves.
_The transmitted light that changes stepwise up to ₂ will be received in time series. And this is the i column (ie (i, 1) ~
The pixels (i, j) to (i, n)) are exactly the same, and the pixel rows in the horizontal direction 1 to i to n that are divided in the vertical direction sequentially receive the same light reception in time series. Will be done.

Then, the signals based on the transmitted light of the wavelength λ ₁ received first by each pixel are integrated to obtain the first spectral image P (λ ₁ ) near 550 nm, and then specified by Δλ. The second spectral image P (λ
₁ + Δλ), and so on, similarly for the third spectral image P (λ ₁ +
2Δλ), kth spectral image (λ ₁ + (k-1) Δλ)
.. and image signals by integrating signals based on the spectrum of each wavelength range are sequentially performed up to the spectral image P (λ ₂ ) of the wavelength λ ₂ , P (λ ₁ ) to P ( A total of p spectral images P of each wavelength up to λ ₂ ) will be obtained. Such image processing can be constructed using known computer technology.

In addition to the above, according to the present invention, a continuous spectral image of the entire measuring object (entire image) can be obtained, and at the same time, a spectral spectrum of each minute portion can be obtained. By obtaining a spectral image, it is possible to obtain accurate color information in the visible light region, which cannot be obtained by a conventional color camera of the approximate color expression method of R (red), G (green), and B (blue). It is possible to utilize, and it is also possible to suppress variations due to a measurement site, unevenness in sensitivity of sensor elements, uneven illumination, etc., which occur due to being a natural product, by averaging and differentiating the spectral spectrum.

[0027]

EXAMPLE A description will be given below of a specific example of the present invention by using an example of a fruit selecting device for evaluating the quality of apples shown in FIGS.

In these figures, 1 is a lighting device, 2 is a light source lamp of the lighting device 1, 3 is an apple to be measured, 4 is a CCD camera, a lens 5, a continuous interference filter 6 and fixed inside thereof. The CCD sensor 7 is a two-dimensional image sensor and the controller 8.

As shown in the figure, a plurality of the light source lamps 2 are provided so that the entire upper surface of the apple 3 stopped on the measuring stage is illuminated, and, for example, a halogen lamp or an infrared light bulb is used.

The CCD camera 4 is fixedly supported by a fixed mounting frame 9 so that the apple 3 can be imaged by the conveyor 10 at a constant speed and stopped on the measuring stage. And the CCD camera 4 uses the light source lamp 2
The reflected light from the apple 3 illuminated by the lens 5,
The CCD sensor 7 receives and forms an image through the continuous interference filter 6. As a result, a fixed positional relationship between the apple 3 stopped at the measuring stage and the CCD sensor 7 fixed inside the illumination optical system and the CCD camera is maintained.

The continuous interference filter 6 and the CCD sensor 7
Is the same as that described in FIGS. 1 and 2 above. That is, the continuous interference filter 6 of this example has predetermined wavelength ranges λ _{1 to} λ ₂ , for example, 400 to 700 nm (visible light region), 700 to 2500 nm (near infrared region), 500 to
1000 nm (over both areas above), 2500-300
Within a wide range such as 0 nm (a part of the infrared range), for example, a narrow region (one specific wavelength) is subdivided so that the wavelength of the transmitted light changes stepwise at a predetermined step of about several tens nm. Areas corresponding to the band width are connected in the left-right direction of FIG. 3 (the rotation direction of the conveyor 10), while the transmitted light wavelength is constant in the direction orthogonal to this (the depth direction of FIG. 3). Which is arranged in the left-right direction with respect to the CCD sensor 7 fixed inside the CCD camera 4,
It is moved at a predetermined intermittent (a period of time required to move one specific wavelength band width and a stop time required to charge the CCD sensor with transmitted light). Further, the CCD sensor 7 arranged above the continuous interference filter 6 so as to face it has m × n pixels, and
The transmitted light of a predetermined wavelength that has passed through the subdivided wavelength range is received by each pixel column (there is m column) arranged in the depth direction. That is, one pixel row receives transmitted light of the same wavelength, and the wavelength of the transmitted light received by each row passes through which of the wavelength ranges of the continuous interference filter 6 facing each other changes stepwise. It depends on whether or not

Reference numeral 11 in FIG. 3 is a detector for detecting that the apple 3 placed on the tray 13 and conveyed has reached the measuring stage. The detection signal is used to control the image pickup of the CCD camera 4. The apples 3 are sent to the control unit 8 and are sent to a controller 14 described later to stop the driving of the conveyor 10 or the apple conveyance by a stopper (not shown) so that the apples 3 are stopped on the measuring stage. It has become.

Reference numeral 12 is a pulse signal transmitter provided in the driving (driven) pulley portion of the conveyor 10, and is provided so as to send the traveling information of the conveyor 10 to the controller 8 and the controller 14.

Next, the operation contents of the image spectroscopy performed in this example and the spectral image obtained by this will be described with reference to FIGS. Note that FIG. 4 shows the apparatus of this example by a schematic block for convenience of description, and therefore the arrangement relationship of each configuration does not necessarily match with FIG. 3, and similarly, FIG. 6 is connected to the CCD sensor 7. In order to easily understand the relationship between the imaged image and the transmitted light due to the movement of the continuous interference filter 6, these positional relationships are shown by being changed by 90 °.
Actually, these are in a positional relationship in which the flat plates are parallel to each other.

Now, in FIG. 4, the CCD camera 4 is
The CCD sensor 31, which is fixed inside so that the cooling device 31 can be easily assembled for noise reduction, is moved up and down in FIG. 4 by the DC motor 32 (left and right in FIG. 3).
The continuous interference filter 6 is movably supported by a support body and a movement guide mechanism (not shown). Reference numeral 30 denotes, for example, a Pertu cooling element (electronic cooling), which circulates cold water between the cooling water circulating device 31 and heat exchange to cool the CCD sensor.

The continuous interference filter 6 is moved by the DC motor 32 so as to repeat the movement of one specific wavelength band width and the stop necessary for charging the CCD sensor with the transmitted light. The timing of the movement start is given by the detector 11 detecting the transfer of the apple 3 to the measurement stage. The drive of the DC motor 32 is controlled by the local microprocessor 33. In addition,
The continuous interference filter 6 is, of course, provided with appropriate light-shielding means except the light transmitting surface so that light does not enter the CCD sensor 7 in the system path other than the filter transmitting surface. is there. CC through the continuous interference filter 6
The electric charge received by the D sensor 7 and charged in each pixel in proportion to the amount of received light becomes image data, which is read from the CCD sensor 7 for each pixel at a predetermined timing in synchronization with the movement timing of the filter 6. Information stored in the memory board 35 and necessary for image processing of the spectral image is sent from the control unit 8 to the central processing circuit 36 of the control device 14.

FIG. 5 shows a case where the movement of the continuous interference filter 6, the image pickup by the CCD sensor 7, and the reading of the image data are performed for one apple 3.

FIG. 6 shows the relationship between the image of the apple 3 formed on the CCD sensor 7 and the wavelength of the transmitted light received by each pixel as the continuous interference filter 6 moves. Yes, (a) shows a state in which the light transmitted through the region of wavelength λ ₁ is received by the pixel row (the shaded portion in the figure) at the position where the continuous interference filter 6 has moved to the third step, Similarly, (B) shows the state of moving to the fourth step, and (C) shows the state of moving to the ninth step.

Focusing only on the pixel row at the third step, for example, in the case of (b), the light transmitted through the region of wavelength λ ₁ + Δλ is received by the pixel row at the third step, In the case of (c), the light transmitted through the region of wavelength λ ₁ + 8 · Δλ is received by the pixel row of the third step. In this way, in one pixel row, the light in each region of wavelengths λ _{1 to} λ ₂ is sequentially received with the constant intermittent movement of the continuous interference filter 6, and the charges charged by the received light (that is, spectrally separated). The image data of the entire CCD sensor 7 can be obtained by reading each image data). FIG. 7 is a table showing a memory state of image data for each wavelength region of each pixel obtained in this way.

From the data based on the spectrally transmitted light of each pixel obtained as described above, for example, the wavelength λ ₁
By integrating data of all the pixels relating, to obtain spectral image P (lambda ₁₎ shown in FIG. 2, all the spectral image by to data integration of all pixels for each wavelength in the same manner to obtain the respective separately It will be.

Separately from the spectroscopic image, if only the data of each wavelength region of the specific pixel (i, j) in the CCD sensor 7 is integrated, as shown in FIG. A spectral spectrum of the received light can be obtained.

An example of apple selection performed by incorporating the above image spectral data into the central processing circuit 36 of FIG. 4 is performed according to FIG. 3, and in the control device 14 of FIG. D conversion means l5, buffer memory l6, measurement calculation means l7 for comparing with signals from the standard value setting means 20, memory l8 for storing image data, reading means l
9, and the arithmetic means 21 obtains images at a plurality of specific wavelengths that are closely related to the quality and characteristics of fruits and vegetables,
The average value for each pixel is read out from the memory 18 via the reading means 19 and the calculation result is sent to the judging means 22.

Next, in this judging means 22, the arithmetic unit 2
From each calculation result sent from No. 1, the quality and characteristics of the apple 3 are judged based on the data (model formula etc.) relating to the quality and characteristics of the fruits and vegetables set in the setting means 23, and the judgment results are used as the sorting control means. 24, and the discharging device 26 is operated to discharge the apple 3 together with the tray 13 onto the take-out conveyor 25. Known methods and devices may be used for this sorting control and discharge control.

If apples are selected as described above, there is an effect that excellent selection can be performed efficiently and quickly by utilizing the information of the spectral image or the spectral spectrum of each pixel. .

[0045]

According to the image spectroscopic method of the present invention, excellent information for quality evaluation of fruits and vegetables can be obtained, and the image spectroscopic device used for this purpose has a simple and compact structure. There is an effect that can be done.

Further, according to the image spectroscopy method of the present invention, a spectral image of the entire image can be obtained, and at the same time, a spectrum of each pixel can be obtained. Therefore, for example, based on the transmitted light of fruits and vegetables, the internal quality (sugar content) , Internal troubles, etc.) can be inspected for each minute portion.

Furthermore, it is possible to quickly obtain a large number of spectral images. Therefore, for example, in the field of quality evaluation and selection of fruits and vegetables, while continuously conveying a large number of fruits on a conveyor, Alternatively, since it can be evaluated quickly and with high efficiency while stopped at the measurement stage, it is the first time in the quality evaluation of fruits and vegetables on an actual industrial scale that the selection based on the internal quality of fruits and vegetables by the spectral image method using transmitted light is performed. And so on.

Further, while the two-dimensional image sensor and the object to be measured are fixed with respect to the illumination system, only by moving the continuous interference filter in one direction, each pixel is made to receive the spectrum of the changing wavelength in time series. Therefore, there is an effect that imaging can be performed without causing a change in optical conditions between the measurement target and individual pixels, and a highly accurate spectral image can be obtained.

[Brief description of drawings]

FIG. 1 is a view for explaining the principle of image spectroscopy of the present invention,

FIG. 2 is a diagram showing a spectral image obtained by the image spectroscopy of the present invention,

FIG. 3 is a view showing an outline of an apparatus when the present invention is applied to quality inspection of apples,

FIG. 4 is a block diagram showing a schematic configuration of an image spectroscopic device portion in FIG. 3;

FIG. 5 is a flow chart for explaining the movement relationship between the continuous interference filter and the operation of the two-dimensional image sensor,

FIG. 6 is a diagram for explaining the imaging relationship between each pixel of the measurement target and the λ ₁ region of the continuous interference filter;

FIG. 7 is a diagram showing a relationship between each pixel and a signal to be read from the pixel,

FIG. 8 is a diagram showing a spectrum of a pixel (i, j) point;

FIG. 9 is a diagram for explaining an example of a conventional image spectroscopy method.

[Explanation of symbols]

1: two-dimensional image sensor, 2: continuous interference filter,
3: apple, 4: CCD camera, 5: lens, 6: continuous interference filter, 7: CCD sensor, 8: control unit, 10: conveyor, 11: detector, 12: pulse signal transmitter, 14: control device.

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] November 10, 1993

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] Brief description of the drawing

[Correction method] Change

[Correction content]

[Brief description of drawings]

FIG. 1 is a view for explaining the principle of image spectroscopy of the present invention,

FIG. 2 is a diagram showing a spectral image obtained by the image spectroscopy of the present invention,

FIG. 3 is a view showing an outline of an apparatus when the present invention is applied to quality inspection of apples,

FIG. 4 is a block diagram showing a schematic configuration of an image spectroscopic device portion in FIG. 3;

FIG. 5 is a flow chart for explaining the movement relationship between the continuous interference filter and the operation of the two-dimensional image sensor,

Figure 6 (a), (b), (c) is a diagram for explaining an imaging relationship lambda ₁ region of a continuous interference filter and the pixels of the measurement object,

Figure 7, each pixel, Chart showing the relationship of signals will now be read,

FIG. 8 is a diagram showing a spectrum of a pixel (i, j) point;

FIG. 9 is a diagram for explaining an example of a conventional image spectroscopy method.

[Explanation of Codes] 1: Two-dimensional image sensor, 2: Continuous interference filter,
3: apple, 4: CCD camera, 5: lens, 6: continuous interference filter, 7: CCD sensor, 8: control unit, 10: conveyor, 11: detector, 12: pulse signal transmitter, 14: control device.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Kaji Matsumoto 630 Shinogase Town, Hamamatsu City, Shizuoka Prefecture

Claims (5)

[Claims] 1. A small area through which only light of a specific wavelength is transmitted is provided by connecting a plurality of narrow areas divided in one direction of a surface having a two-dimensional spread, and each of the connected narrow areas is specified. A continuous interference filter provided so that the transmission wavelength of the light continuously or stepwise changes along the one direction, is relatively moved in the one direction between the measurement object and the two-dimensional image sensor, An image spectroscopic method characterized in that each pixel of a two-dimensional image sensor sequentially detects light from an object to be measured, as filter-transmitted light whose wavelength continuously or stepwise changes, for each wavelength. 2. An image spectroscopic method, wherein signals based on a common transmission wavelength detected by each pixel of claim 1 are integrated to create a spectral image for each wavelength. 3. The image spectroscopic method according to claim 1, wherein the transmission wavelength of the light transmitted by the continuous interference filter is in the range of visible light region to near-infrared light region. 4. A large number of narrow regions, which allow only light of a specific wavelength to pass therethrough, are formed by dividing the sections into one direction on a surface having a two-dimensional spread, and connecting each of the narrow regions. A continuous interference filter which is provided so that the transmission wavelength of light continuously or stepwise changes along the one direction, and one surface of which faces a measurement target, and a continuous interference filter on the opposite side of the continuous interference filter. A two-dimensional image sensor arranged to face the surface, a filter moving means for relatively moving the continuous interference filter in the one direction with respect to the measurement object and the two-dimensional image sensor, and transmission through the moving continuous interference filter. Signal reading means for sequentially reading from the two-dimensional image sensor as a signal obtained by photoelectrically converting filter-transmitted light received by each pixel of the two-dimensional image sensor,
An image spectroscopic device comprising: 5. The method according to claim 4, wherein signals based on a common transmission wavelength among signals read from each pixel are integrated,
An image spectroscopic device comprising image processing means for creating spectral images for each of the divided transmission wavelengths.