JP6403872B2 - Fruit and vegetable inspection equipment - Google Patents

Description

  The present invention relates to a fruit and vegetable inspection apparatus for inspecting the presence or absence of abnormality in the fruit skin surface and inside the fruit skin, such as water rot appearing on the citrus fruit skin surface or abnormal drying of the fruit skin.

  In citrus fruits, when the skin is damaged, bacteria enter the skin and the surface of the skin becomes wet for a long time due to rain, dew, etc. Also, water rot disease under conditions of ambient temperature around 25 degrees Symptoms called may appear.

  As specific symptoms of water rot, the citrus peel becomes swollen, and the rot spreads from the generation site to a wide area of the peel, causing mold or dry rot.

  For this reason, when citrus fruits with water rot are mixed and packed with normal products, there is a risk of rot to normal products, so in order to ensure the quality of the product, It is desired to eliminate it at the selection stage.

  For the wounds of the skin, which can be said to be the initial symptom of water rot, especially for wounds immediately after the occurrence, as disclosed in Patent Documents 1 and 2, a specific fluorescence wavelength derived from a substance emitting fluorescence in the visible region is detected by ultraviolet irradiation. Image inspection is performed.

  By detecting such scratches on the skin, which are the initial symptoms of water rot, mold and dry rot, it is possible to prevent defective products from leaking to some extent in the fruit selection process. In the case of fruits and vegetables that have rotted, the skin swells as water rot progresses, so that the skin surface scratches are hidden, and water rot has occurred in the image inspection disclosed in Patent Documents 1 and 2. Fruits and vegetables may not be detected.

  In addition, as a method for inspecting the presence of water rot, mold and dry rot, for example, as disclosed in Patent Document 3, the fruit and vegetables are irradiated with light from an illumination lamp such as a halogen lamp, There is known a method for detecting the presence or absence of a discolored or corrupt portion of fruit and vegetables by imaging reflected light with an imaging camera.

  Further, as disclosed in Patent Document 4, the light for inspection is irradiated from below the fruits and vegetables by the light projecting means, and the fruits and vegetables are picked up by the CCD camera disposed above, and the three primary colors (R, R, G, B) are used to obtain the difference value between the R signal, in which the difference in the amount of transmitted light is likely to appear as a difference in the image signal, and the G signal, B signal, which are difficult to appear, depending on the presence or absence of the decayed portion. Based on this, a method for detecting the presence or absence of a rot portion is also known.

Japanese Patent Laid-Open No. 2003-14650 JP 2011-33612 A Japanese Patent Laid-Open No. 9-24343 JP 2006-10511 A

  However, in the method disclosed in Patent Document 3, since the inspection is performed using the saturation, chromaticity, and brightness of the image captured by the imaging camera, a defect that causes a change in visible light such as citrus blue mold. However, when there is almost no difference in saturation, chromaticity, and lightness between normal and rotting parts, such as water rot and dry rot, it is difficult to make an accurate determination. May be judged as a normal product.

  Further, in the water rot detection method disclosed in Patent Document 4, an image based on visible light is taken by a CCD camera, and an R signal included in the image has an average light amount in a wavelength range of about 550 nm to 700 nm. It is only detecting.

  For this reason, the sensitivity is low to detect minute water rot or dry rot of about 10 mm in diameter as required in the recent market, and sufficient inspection cannot be performed.

  In the present invention, in view of the current situation, fruits and vegetables that can accurately detect abnormalities such as rot and scratches existing on the surface and inside of the skin of fruits and vegetables, and can detect even minute water rot. An object is to provide an inspection device.

The present invention was invented in order to solve the problems in the prior art as described above.
A fruit and vegetable inspection apparatus for determining the presence or absence of abnormality of fruit and vegetables,
A light projecting means for irradiating the fruit and vegetables with inspection light;
Imaging means for imaging the fruits and vegetables with the inspection light;
Analysis means for detecting the presence or absence of abnormality of the fruits and vegetables based on the inspection image of the fruits and vegetables imaged by the imaging means,
The light projecting means can be irradiated with light including at least the absorption wavelength of water,
The analysis means is configured to detect the presence or absence of abnormality of the fruits and vegetables using an inspection image based on the light having an absorption wavelength of water.

  In this case, it is preferable that the imaging means uses an InGaAs photodiode as an imaging element.

  Further, the imaging means can take an inspection image of the fruits and vegetables with reflected light obtained by reflecting the inspection light emitted from the light projecting means on the fruits and vegetables.

  Further, the imaging means can take an inspection image of the fruits and vegetables with transmitted light obtained by transmitting the inspection light emitted from the light projecting means through the fruits and vegetables.

  Moreover, in the fruit and vegetable inspection apparatus of this invention, the disorder | damage | failure related to the increase / decrease in the water | moisture content which appears on the fruit skin surface layer and / or under the fruit skin as said fruit and fruit abnormality can be detected.

  According to the present invention, by using an inspection image based on light having an absorption wavelength of water, even in the case of water rot of fruits and vegetables, which is difficult to discriminate in an image based on visible light, it is clear that the normal portion and the rotting portion. Contrast is generated and can be easily and accurately determined.

FIG. 1 is a schematic configuration diagram for explaining the configuration of an embodiment of the fruit and vegetable inspection apparatus of the present invention. FIG. 2 (a) is a grayscale image obtained by imaging the fruits and vegetables S with visible light, and FIG. 2 (b) is an inspection image based on 1200 nm light as the water absorption wavelength for the fruits and vegetables S in FIG. 2 (a). It is an analysis image at the time of comparing with the inspection image based on the light of 1030 nm as light of the predetermined absorption wavelength used as a standard. 3 (a) is a grayscale image obtained by imaging the fruits and vegetables S with visible light, FIG. 3 (b) is an inspection image based on light of 1160 nm as the water absorption wavelength for the fruits and vegetables S in FIG. 3 (a), It is an analysis image at the time of taking the difference with the test | inspection image based on the light of 1135 nm as light of the predetermined | prescribed absorption wavelength used as a reference | standard. 4A is a grayscale image obtained by imaging the fruits and vegetables S with visible light, and FIG. 4B is an inspection image 2 based on 1200 nm light as the water absorption wavelength for the fruits and vegetables S in FIG. 4A. It is the analysis image which took the second derivative. FIG. 5 shows spectrum data obtained by measuring normal fruits and vegetables, fruits and vegetables having water rot, and fruits and vegetables having dry rot using the fruit and vegetable inspection apparatus shown in FIG. FIG. 6 is a schematic configuration diagram for explaining a configuration in another embodiment of the fruit and vegetable inspection apparatus of the present invention. FIG. 7 is a schematic configuration diagram for explaining a configuration in still another embodiment of the fruit and vegetable inspection apparatus of the present invention. 8A and 8B show a test for mandarin oranges having mold as the fruits and vegetables S. FIG. 8A shows a visible image and FIG. 8B shows an inspection image. FIG. 9 shows an inspection of mandarin oranges having dried scratches on the surface of fruits and vegetables S. FIG. 9A shows a visible image and FIG. 9B shows an inspection image. FIG. 10 shows an inspection of peaches with so-called press marks as fruits and vegetables S. FIG. 10 (a) shows a visible image, FIG. 10 (b) shows an inspection image, and FIG. 10 (c) peels the skin. It is the visible image of the fruit and vegetables S made into the state which can confirm the inside of a fruit skin. FIG. 11 is a person who inspected pears with water fruits as fruits and vegetables S, FIG. 11 (a) is a visible image, FIG. 11 (b) is an inspection image, and FIG. 11 (c) is a peeled peel. It is the visible image of the fruit and vegetables S which made the state which can confirm the inside.

  Hereinafter, embodiments (examples) of the present invention will be described in more detail with reference to the drawings.

  FIG. 1 is a schematic configuration diagram for explaining the configuration of an embodiment of the fruit and vegetable inspection apparatus of the present invention.

  As shown in FIG. 1, the fruit and vegetable inspection apparatus 10 of the present embodiment includes a light projecting unit 12 that irradiates inspection light onto the fruit and vegetable S to be measured, and an inspection light (reflected light) reflected by the fruit and vegetable S. Imaging means 14 for picking up images of the fruits and vegetables S, and analysis means 16 for detecting the rot portion of the fruits and vegetables S based on the inspection image of the fruits and vegetables S taken by the imaging means 14.

  In the present embodiment, the fruits and vegetables S are not particularly limited, but can be, for example, citrus fruits such as mandarin oranges and citrus fruits, pears, peaches, loquats, plums, and apples.

  Further, in the case of such fruits and vegetables S, in the fruits and vegetables inspection apparatus 10 of this embodiment, for example, water rot found in citrus fruits, water fruits found in pears, peaches, loquats, plums, apples, etc. It is possible to inspect the press marks.

  Here, “water rot” means that when the skin is damaged, as described above, bacteria enter the wound, and the surface of the skin is wet for a long time due to rain or dew. It is a symptom that appears as if the pericarp swells, appearing under temperature conditions.

  In addition, “water fruit” is a symptom in which the pulp is immersed in water. When the degree becomes severe, the pulp becomes brownish.

  In addition, “pressing marks” are the result of local pressure being applied to the surface of fruits and vegetables by contact between the fruits and vegetables, destroying the flesh tissue of the fruits and vegetables, and moisture exuded from the flesh tissue between the peel and the flesh. This is a symptom in which a state to perform (in other words, a state of internal bleeding in the human body) appears.

  The fruit and vegetable inspection apparatus 10 according to the present embodiment is not limited to the inspection of such a failure, but, for example, is a general failure related to the increase or decrease of water that appears in the skin surface and / or under the skin when the pulp cell is destroyed. It is possible to check for.

  The light projecting means 12 is not particularly limited as long as it is near infrared light of 900 nm to 2000 nm and can irradiate light including the absorption wavelength of water. For example, a halogen lamp or LED light source is used. Can be used. The LED light source may be one that emits white light, but may be one that emits only light of a specific wavelength.

  In addition, although the absorption wavelength of water is known as 960 nm, 1150 nm, 1450 nm, and 1940 nm, since the absorption wavelength of water is not a specific wavelength but exists as a wide wavelength band, if absorption by water can be confirmed A wavelength slightly around may be used.

  The imaging unit 14 is not particularly limited as long as it can capture an image based on the inspection light having the wavelength irradiated by the light projecting unit 12, and is not limited to an area camera, a line camera, an imaging spectrometer, and a multiband camera. Etc. can be used. In particular, it is preferable to use a photodiode such as InGaAs, Ge, or PbS that can detect near-infrared light of 900 nm to 2000 nm as the imaging device of the imaging unit 14.

  Note that a band-pass filter 18 that transmits only light of a predetermined wavelength may be provided between the fruit and vegetables S and the imaging unit 14.

  With this configuration, the imaging unit 14 can receive only light having a necessary wavelength and does not receive light having a wavelength unnecessary for image analysis, so that noise can be reduced.

  In the fruit and vegetable inspection apparatus 10 of the present embodiment, the fruit and vegetables S are irradiated with inspection light from the light projecting means 12 and the fruit and vegetables S are picked up by the imaging means 14 using the reflected light from the fruits and vegetables S to obtain an inspection image. doing.

  Each pixel value of the inspection image can be determined based on the light amount L of the inspection light received by the imaging unit 14, but in this embodiment, the reflected light from the fruits and vegetables S is expressed by the following equation (1). And each pixel value of the inspection image based on the reflection ratio of the fruits and vegetables S calculated as a ratio to the reflected light from a standard body (for example, a gray chart) obtained by irradiating incident light acquired in advance. Has been decided. In addition, as represented by the following formula (2), the pixel value may be determined based on the apparent absorbance from the calculated reflection ratio.

  In this way, by using the reflected light Rr from the standard as a reference, the reflection ratio R can be measured with almost no variation even when the light amount is reduced due to, for example, deterioration of the light projecting means 12. Therefore, a stable inspection can be performed for a long time.

  The determination of each pixel value of the inspection image based on the reflection ratio R and the apparent absorbance A can be performed as follows, for example.

  For example, in the case of an 8-bit image, the pixel value is a value from 0 to 255. Therefore, the assumed minimum value of the reflection ratio R (appropriately set based on the performance of the imaging unit 14) is “0”, and the reflection ratio R is The reflection ratio R of each pixel may be converted so that 1 which is the maximum value becomes “255”.

  The inspection image is analyzed by the analysis unit 16 as will be described later, so that the rot portion of the fruits and vegetables S can be detected.

  In the image analysis in the analysis means 16, (1) the ratio of inspection images based on light of a plurality of wavelengths (2) difference of inspection images based on light of a plurality of wavelengths (3) inspection image based on light of absorption wavelength of water By performing any of the second order differentiation, an analysis image in which the corruption site in the inspection image is clarified is generated, and the corruption site is detected.

  In addition, these image analyzes, because the amount of water in the rot portion of the fruits and vegetables S is larger than that in the normal portion, the light of the absorption wavelength of water is absorbed by the rot portion, and when imaged by the imaging means 14, This is based on the fact that the amount of light at the decayed portion is reduced as compared with the normal portion.

  Hereinafter, each image analysis will be described in detail.

(1) In the ratio of inspection images based on light of a plurality of wavelengths, an inspection image based on light having a water absorption wavelength λ ₁ and an inspection image based on light having a predetermined absorption wavelength λ ₂ as a reference are used. As shown in the following formula (3), the decaying part is specified by taking a light quantity ratio for each pixel.

  Here, X is any one of the light quantity L, the reflection ratio R, and the apparent absorbance A.

  FIG. 2 shows an example of an analysis image when the ratio of inspection images is taken.

  FIG. 2 (a) is a grayscale image obtained by imaging the fruits and vegetables S with visible light, and FIG. 2 (b) is an inspection image based on 1200 nm light as the water absorption wavelength for the fruits and vegetables S in FIG. 2 (a). It is an analysis image at the time of taking a ratio with the inspection image based on the light of 1030 nm as light of the predetermined absorption wavelength used as a standard.

  As shown in FIG. 2A, when imaging is performed with visible light, there is almost no difference in saturation, chromaticity, and lightness between the normal site X and the rotting site Y, but as shown in FIG. Thus, by comparing the inspection images, a clear contrast is generated between the normal site X and the rot site Y, and the presence or absence of the rot site Y can be easily and reliably determined.

(2) In the difference between the inspection images based on the light having a plurality of wavelengths, the inspection image based on the light having the water absorption wavelength λ ₁ and the inspection image based on the light having the reference absorption wavelength λ ₂ are used. As shown in the following formula (4), the decaying part is specified by taking a difference in light quantity for each pixel.

  Here, X is any one of the light quantity L, the reflection ratio R, and the apparent absorbance A.

  FIG. 3 shows an example of an analysis image when a difference between inspection images is taken.

  3 (a) is a grayscale image obtained by imaging the fruits and vegetables S with visible light, FIG. 3 (b) is an inspection image based on light of 1160 nm as the water absorption wavelength for the fruits and vegetables S in FIG. 3 (a), It is an analysis image at the time of taking the difference with the test | inspection image based on the light of 1135 nm as light of the predetermined | prescribed absorption wavelength used as a reference | standard.

  As shown in FIG. 3A, when imaging is performed with visible light, there is almost no difference in saturation, chromaticity, and lightness between the normal site X and the rotting site Y, but as shown in FIG. Thus, by comparing the inspection images, a clear contrast is generated between the normal site X and the rot site Y, and the presence or absence of the rot site Y can be easily and reliably determined.

  Further, in the second derivative of the inspection image, by using the inspection image based on the light of the absorption wavelength of water and the inspection image based on the light of the wavelength before and after that, by taking the second derivative of the light amount for each pixel, The site of corruption is identified.

  An approximate expression can be used for the calculation of the second derivative.

Specifically, the inspection image B based on light absorption wavelength lambda _B, the inspection image A than the absorption wavelength lambda _B of water based on light of a predetermined wavelength shorter wavelength lambda _A, longer predetermined wavelength than the absorption wavelength lambda _B The analysis image D can be obtained by performing the calculation of the following formula (5) for each pixel using the inspection image C based on the light of the wavelength λ _C.
P _D = P _A −2 × P _B −P _C (5)
In the above formula (5), P _A : pixel signal P _B of the inspection image A: pixel signal P _C of the inspection image B: pixel signal P _D of the inspection image C: pixel signal of the analysis image D

  By performing the calculation for each pixel in this manner, an analysis image obtained by taking the second derivative of the inspection image based on the light having the absorption wavelength of water can be generated.

  In the present embodiment, in order to simplify the description, a single light projecting unit 12 and a single image capturing unit 14 are provided. However, a plurality of light projecting units 12 may be provided. In addition, a plurality of imaging means 14 may be provided.

  FIG. 4 shows an example of an analysis image when taking the second derivative of the inspection image.

  4A is a grayscale image obtained by imaging the fruits and vegetables S with visible light, and FIG. 4B is an inspection image 2 based on 1200 nm light as the water absorption wavelength for the fruits and vegetables S in FIG. 4A. It is the analysis image which took the second derivative.

  As shown in FIG. 4A, when imaging is performed with visible light, there is almost no difference in saturation, chromaticity, and lightness between the normal site X and the rotting site Y, but FIG. As shown, by comparing the inspection images, a clear contrast occurs between the normal site X and the rot site Y, and the presence or absence of the rot site Y can be easily and reliably determined.

  In addition, although the present Example demonstrated based on an example about what the water rot produced as abnormality of fruit and vegetables, even if it is a case where dry rot has arisen in the fruit and skin of fruit and vegetables, a spectrum different from a normal site | part is obtained. Therefore, it is possible to detect abnormalities in fruits and vegetables by performing image analysis in the same manner.

  FIG. 5 shows spectrum data obtained by measuring normal fruits and vegetables, fruits and vegetables having water rot, and fruits and vegetables having dry rot using the fruit and vegetable inspection apparatus 10 of FIG. Note that the spectral data shown in FIG. 5 is obtained by subjecting absorbance to second-order differentiation processing at intervals of a wavelength of 25 nm.

  As shown in FIG. 5, the fruits and vegetables with water rot show an increase in absorption at around 960 nm and around 1150 nm, which are the absorption wavelengths of water, compared to normal fruits and vegetables.

  Moreover, about the fruits and vegetables which have dry rot, compared with normal fruits and vegetables, the fall of absorption is seen in the water absorption wavelength around 960 nm and around 1150 nm.

  That is, it can be seen that abnormalities such as water rot of fruits and vegetables can be detected by using a test image based on light having an absorption wavelength of water and observing a change in the degree of absorption of the test light.

  On the other hand, when abnormal drying occurs in the fruit and fruit skin, light having an absorption wavelength of water is not absorbed as compared with normal fruit and vegetables. For this reason, it is possible to discriminate between normal fruits and vegetables and fruits and vegetables whose skins are abnormally dried.

  FIG. 6 is a schematic configuration diagram for explaining a configuration in another embodiment of the fruit and vegetable inspection apparatus of the present invention.

  The fruit and vegetable inspection apparatus 10 shown in FIG. 6 has basically the same configuration as the fruit and vegetable inspection apparatus 10 shown in FIGS. 1 to 5, and the same components are denoted by the same reference numerals and detailed description thereof is omitted. To do.

  In the fruit and vegetable inspection apparatus 10 shown in FIG. 1, the light projecting means 12 and the imaging means 14 are arranged in the same direction with respect to the fruit and vegetables S, and the fruit and vegetables S are imaged by reflected light. Then, the inspection light irradiated by the light projecting means 12 passes through the fruits and vegetables S, and an inspection image of the fruits and vegetables S is picked up by the imaging means 14 using the transmitted light.

  Thus, even when the inspection image based on the transmitted light is used, as described above, by performing image analysis as in the case of using the inspection image based on the reflected light, the normal part of the fruits and vegetables S And rot sites can be discriminated.

  In this embodiment, the inspection image is picked up only by the transmitted light. However, the inspection image may be picked up by using both the transmitted light and the reflected light in combination with the above-described embodiment.

  FIG. 7 is a schematic configuration diagram for explaining a configuration in still another embodiment of the fruit and vegetable inspection apparatus of the present invention.

  The fruit and vegetable inspection apparatus 10 shown in FIG. 7 has basically the same configuration as the fruit and vegetable inspection apparatus 10 shown in FIGS. 1 to 6, and the same components are denoted by the same reference numerals and detailed description thereof is omitted. To do.

  The fruit and vegetable inspection apparatus 10 shown in FIGS. 1 to 6 is configured such that inspection light is irradiated from the light projecting unit 12 to the stationary fruit and vegetable S, and an inspection image based on the inspection light is captured by the imaging unit 14. However, in the fruit and vegetable inspection apparatus 10 of this embodiment, the inspection light is irradiated to the fruit and vegetables S conveyed in one direction by the conveying means 20 and an inspection image is taken.

  In this way, a large amount of fruits and vegetables can be efficiently inspected by configuring so as to inspect the fruits and vegetables in-line.

  When the inspection is performed while the fruits and vegetables S are transported by the transport means 20, as shown in FIG. 7, the side surfaces of the fruits and vegetables S are reflected on the reflectors by providing the reflecting mirrors 22 on both sides in the transport direction. Thus, it is preferable that the entire image of the fruits and vegetables S be imaged by the imaging unit 14.

  8 to 12 show examples of visible images and inspection images when the fruits and vegetables S are inspected by using the fruits and vegetables inspection apparatus 10 shown in FIG.

  8A and 8B show a test for mandarin oranges having mold as the fruits and vegetables S. FIG. 8A shows a visible image and FIG. 8B shows an inspection image.

  Mold appearing at the top in the visible image can be confirmed to be white in the inspection image.

  FIG. 9 shows an inspection of mandarin oranges having dried scratches on the surface of fruits and vegetables S. FIG. 9A shows a visible image and FIG. 9B shows an inspection image.

  Dry scratches appearing at the bottom of the visible image can be confirmed to be white in the inspection image.

  When the moisture on the skin surface or under the skin is reduced, such as the mold in FIG. 8 and the dry scratch in FIG. 9, the absorption rate of near-infrared light is low. Appears white.

  FIG. 10 shows an inspection of peaches with so-called press marks as fruits and vegetables S. FIG. 10 (a) shows a visible image, FIG. 10 (b) shows an inspection image, and FIG. 10 (c) peels the skin. It is the visible image of the fruit and vegetables S made into the state which can confirm the inside of a fruit skin.

  A pressing mark that is difficult to confirm in the visible image (a portion where the color is dark in FIG. 10C) can be confirmed black in the inspection image.

  FIG. 11 is a person who inspected pears with water fruits as fruits and vegetables S, FIG. 11 (a) is a visible image, FIG. 11 (b) is an inspection image, and FIG. 11 (c) is a peeled peel. It is the visible image of the fruit and vegetables S which made the state which can confirm the inside.

  Water fruits that are difficult to confirm in the visible image (locations where the color is dark in FIG. 11C) can be confirmed black in the inspection image.

  When the moisture on the skin surface or under the skin is increased, such as the pressing mark in FIG. 10 and the water fruit in FIG. 11, the absorption rate of near-infrared light is high. Appears black.

  The preferred embodiment of the present invention has been described above, but the present invention is not limited to this. For example, an inspection image based on light of two wavelengths is used as a comparison of inspection images based on light of a plurality of wavelengths. However, various modifications are possible without departing from the object of the present invention, such as image analysis using inspection images based on light of three or more wavelengths. is there.

DESCRIPTION OF SYMBOLS 10 Fruit and vegetable inspection apparatus 12 Light projection means 14 Imaging means 16 Analysis means 18 Band pass filter 20 Conveyance means 22 Reflecting mirror

Claims (3)

A fruit and vegetable inspection apparatus for determining the presence or absence of abnormality of fruit and vegetables,
A light projecting means for irradiating the fruit and vegetables with inspection light;
Imaging means for imaging the fruits and vegetables with the inspection light;
Analysis means for detecting the presence or absence of abnormality of the fruits and vegetables based on the inspection image of the fruits and vegetables imaged by the imaging means,
The light projecting means can be irradiated with light including at least the absorption wavelength of water,
The analysis means is configured to detect the presence or absence of abnormality of the fruits and vegetables using an inspection image based on light of the water absorption wavelength.
The imaging means captures an inspection image of the fruits and vegetables with the reflected light reflected from the fruits and vegetables by the inspection light emitted from the light projecting means,
And the reflected light Rs from the previous SL fruits or vegetables, the reflection ratio of the fruits or vegetables which are calculated as the ratio of the reflected light Rr from a standard body that is obtained by irradiating the incident light previously acquired R, i.e., the following equation (1) Calculate
By converting the minimum value of the reflection ratio R to be the minimum value of the pixel value of the inspection image and the maximum value of the reflection ratio R to be the maximum value of the pixel value, each pixel value of the inspection image is converted. A fruit and vegetable inspection apparatus characterized by deciding.
  The fruit and vegetable inspection apparatus according to claim 1, wherein the imaging unit uses a photodiode capable of receiving a wavelength of 900 nm to 2000 nm as an imaging element. The fruit and vegetable inspection apparatus according to claim 1 or 2 , wherein the abnormality of the fruit or vegetable is an obstacle related to an increase or decrease in moisture appearing on a skin layer and / or under the skin of the fruit or vegetable.