KR20240100906A - An apparatus for detecting failure in an object and a method for detecting the failure in the object - Google Patents

Abstract

A method for detecting defects in an object acquires image data for a specific area of the object, determines at least one material based on the acquired image data, and determines whether the determined material is defective.

Description

{An apparatus for detecting failure in an object and a method for detecting the failure in the object}

The embodiment relates to an apparatus for detecting defects in an object and a method for detecting defects in an object.

In the case of manufacturers who produce products based on objects such as raw materials, they want to know the types of substances contained in the raw materials or the characteristics of the substances, but the raw material supplier does not provide such information for reasons such as information security. The type or characteristics of the material contained in the raw material are unknown.

If the manufacturer knows the types of substances contained in the raw materials or the characteristics of the substances, they can determine the causality of how these substances change or are related during the process of manufacturing the product, and based on that causality, what may occur during product production is possible. Defects can be minimized. In addition, if the manufacturer knows the type or characteristics of the material contained in the raw material, information on changing the type of material or adjusting the material properties to minimize defects that may occur during product production is provided to the raw material supplier. By providing feedback, better quality products can be manufactured by receiving better quality raw materials from raw material suppliers.

Therefore, there is a strong demand for technology that can easily identify the type or material characteristics of the material contained in the object.

The embodiments aim to solve the above-described problems and other problems.

Another purpose of the embodiment is to provide a device for detecting defects in an object and a method for detecting defects in an object that can easily identify substances contained in the object.

Another purpose of the embodiment is to provide a device for detecting defects in an object and a method for detecting defects in an object that have a simple structure and can reduce costs by identifying materials based on images obtained from the object.

Another purpose of the embodiment is to provide a device for detecting defects in an object and a method for detecting defects in an object that can easily determine defects in the identified material.

The technical problems of the embodiments are not limited to those described in this item and include those that can be understood through the description of the invention.

According to one aspect of the embodiment to achieve the above or other objects, a method for detecting defects in an object includes acquiring image data for a specific area of the object; determining at least one substance based on the acquired image data; and determining whether the determined material is defective.

The determined materials may include voids, resins, and fillers.

The defect of the determined material may be determined based on at least one of area ratio, size, and density.

The step of determining whether the determined material is defective may include determining whether the void is defective based on the area ratio of the void.

The step of determining whether the determined material is defective may include determining whether the void is defective based on the ferret length of the void.

The step of determining whether the determined material is defective may include determining whether the resin is defective based on the pellet length of the resin.

The step of determining whether the determined material is defective may include determining whether the resin is defective based on the area ratio of the agglomeration area of the resin determined as the density.

The agglomeration area of the resin may be an area less than 0.015 to 1.535% of the AOI size.

The method for detecting defects in the object may include, when the object includes a circuit, performing histogram extreme value correction based on the acquired image data.

According to another aspect of the embodiment to achieve the above or other objects, an apparatus for detecting defects in an object includes: an image acquisition unit that acquires image data for a specific area of the object; a material determination unit that determines at least one material based on the acquired image data; and a defect determination unit that determines whether the determined material is defective.

The determined materials may include voids, resins, and fillers.

The defect of the determined material may be determined based on at least one of area ratio, size, and density.

The defect determination unit may determine whether the void is defective based on the area ratio of the void.

The defect determination unit,

The defect of the void can be determined based on the ferret length of the void.

The defect determination unit may determine whether the resin is defective based on the length of the resin.

The defect determination unit may determine whether the resin is defective based on the area ratio of the agglomeration area of the resin determined as the density.

The agglomeration area of the resin may be an area of 0.015 to 1.535% or less of the AOI size.

The device for detecting defects in an object may include a histogram extreme value correction unit that performs histogram extreme value correction based on the acquired image data when the object includes a circuit.

The effects of the device for detecting defects in an object and the method for detecting defects in an object according to an embodiment will be described as follows.

According to at least one of the embodiments, there is an advantage that various substances contained in the object can be easily identified through image data acquired from the object.

According to at least one of the embodiments, since the material is identified based on the image obtained from the object, there is an advantage in that the structure is simple and the cost can be reduced.

According to at least one of the embodiments, it is possible to easily determine whether each of the at least one determined (identified) material is defective. Accordingly, it is possible to determine not only the type of material contained in the object but also whether the type is defective, making it possible to perform a three-dimensional inspection of the object.

Additional scope of applicability of the embodiments will become apparent from the detailed description that follows. However, since various changes and modifications within the spirit and scope of the embodiments may be clearly understood by those skilled in the art, the detailed description and specific embodiments, such as preferred embodiments, should be understood as being given by way of example only.

1 is a cross-sectional view showing a circuit board according to an embodiment.
Figure 2 is a block diagram showing an apparatus for detecting defects in an object according to an embodiment.
Figure 3 is a flowchart explaining a method for detecting defects in an object according to the first embodiment.
Figure 4 shows detection of substances according to brightness.
Figures 5a to 5f show detection of resin.
Figures 6a and 6b show detection of filler.
Figures 7a and 7b show detection of voids.
Figure 8 shows detection of substances according to color.
Figure 9a shows voids distributed in the AOI image.
Figure 9b is a graph showing defect determination according to the area ratio of voids.
Figure 10 shows how to determine a defect using the ferret length of the void.
Figure 11 shows how to determine defects using the ferret length of the resin.
Figure 12a shows the agglomeration area of the distributed resin in the AOI image.
Figure 12b is a graph showing defective judgment according to the area ratio of the resin agglomeration area.
Figure 13 is a flowchart explaining a method for detecting defects in an object according to a second embodiment.
Figure 14 shows histogram extreme value correction.

Hereinafter, embodiments disclosed in the present specification will be described in detail with reference to the attached drawings. However, identical or similar components will be assigned the same reference numerals regardless of reference numerals, and duplicate descriptions thereof will be omitted. The suffixes 'module' and 'part' for components used in the following description are given or used interchangeably in consideration of ease of specification preparation, and do not have distinct meanings or roles in themselves. In addition, the attached drawings are intended to facilitate easy understanding of the embodiments disclosed in this specification, and the technical ideas disclosed in this specification are not limited by the attached drawings. Additionally, when an element such as a layer, region or substrate is referred to as being 'on' another component, this includes either directly on the other element or there may be other intermediate elements in between. do.

Embodiments may provide a method and device that can easily determine the type of material contained in an object, such as a raw material.

In the following description, a circuit board is representatively presented as a raw material, but embodiments may include raw materials containing at least one or more substances. In addition, the embodiment may include a member containing at least one or more substances in addition to raw materials.

The materials described below may have a shape such as resin, filler, epoxy resin, or glass fabric, or may not have a shape such as a void. Matter and object can be used interchangeably.

1 is a cross-sectional view showing a circuit board according to an embodiment.

Referring to FIG. 1, a circuit board according to an embodiment may include a plurality of prepregs 110, a resin layer 120 between the prepregs 110, and a plurality of vias 151 and 152. .

The prepreg 110 may be formed by impregnating an epoxy resin 111 or the like into a fiber layer in the form of a fabric sheet, such as a glass fabric 112 woven with glass fiber yarn, and then heat-compressing it. However, the embodiment is not limited to this. That is, the prepreg 110 may include a fiber layer in the form of a fabric sheet woven with carbon fiber yarn.

After the resin layer 120 is positioned between the prepregs 110, the substrate can be completed through thermal compression using a press. Thereafter, a via hole is formed by drilling the prepreg 110, and the via hole 151 and 152 can be formed by filling the via hole with a conductive material or plating it with a conductive material. The conductive material may be any one selected from Cu, Ag, Sn, Au, Ni, and Pd. The via hole may be formed by any one of mechanical, laser, and chemical processing.

The vias 151 and 152 may be electrically connected to the circuit patterns 141 to 143 provided on the upper and/or lower side of the prepreg 110.

Meanwhile, the unexplained symbol 113 is a void, which may be an empty space where no material exists. Additionally, the unexplained symbol 130 is a filler. The filler 130 may include various different types of fillers.

Figure 2 is a block diagram showing an apparatus for detecting defects in an object according to an embodiment.

In the following description, for convenience of explanation, the substrate shown in FIG. 1 is limited to the object, but the embodiment can be equally applied to other members other than the substrate shown in FIG. 1.

Referring to FIGS. 1 and 2 , the object defect detection device 200 according to the embodiment includes an image acquisition unit 210, a feature extraction unit 220, a material determination unit 230, and a defect determination unit 240. It can be included.

The image acquisition unit 210 may acquire image data for a specific area of the object.

Image data may be acquired through destructive testing methods. For example, the object may be cut to expose a specific cross-section of the object, and image data may be acquired for the specific cut cross-section. For example, image data may be acquired by, for example, a microscope, SEM or TEM, but there is no limitation thereto.

Meanwhile, in an embodiment, image data for a specific area to be photographed may be acquired through a non-destructive testing method.

The feature extractor 220 may extract a plurality of features included in image data. The plurality of characteristics may include, for example, brightness, size, shape, color, sharpness, etc.

Image data may include various substances that can be distinguished from each other by brightness, border, color, sharpness, etc. In other words, these materials can be distinguished from each other by brightness, border, color, clarity, etc. The substrate shown in FIG. 1 may include a variety of different materials. In other words, the substrate can be formed by composition of various materials. These materials may be resin, glass fabric, filters, etc. Voids are also included in the substrate and can be classified as a type of material.

The material determination unit 230 may determine at least one material based on at least one of the plurality of features extracted by the feature extraction unit 220. That is, at least one material may be determined based on at least one of brightness, size, shape, and color.

Since the substrate shown in FIG. 1 includes resin, at least one filler, glass fabric, and voids, the resin, at least one filler, glass fabric, and voids are determined based on at least one of brightness, size, shape, and color. At least one of these can be determined.

For example, the material determination unit 230 may determine at least one material based on brightness.

For example, the material determination unit 230 may determine at least one material based on brightness and size.

For example, the material determination unit 230 may determine at least one material based on brightness, size, and shape.

For example, the material determination unit 230 may determine at least one material based on brightness, size, shape, and color.

Meanwhile, the material determination unit 230 may determine voids based on clarity. As shown in FIG. 7A, when viewed in color in the image data, both the resin and the void appear black, so they may not be distinguished from each other, and an error may occur when determining the void and the resin. Therefore, considering clarity as a characteristic, resin and voids can be distinguished through the size of clarity. By increasing clarity, the boundaries of voids can be clearly revealed. In this case, considering that the void is relatively very small compared to the resin, a material with a contour angle that is much smaller than the resin in size by increasing clarity can be determined to be a void.

Meanwhile, the defect determination unit 240 may determine defects in at least one of the determined materials. Depending on whether each of the at least one material satisfies preset conditions, it can be determined whether the material is defective.

Meanwhile, the apparatus 200 for detecting defects in an object according to an embodiment includes a storage unit (not shown) that stores image data and various information for each material, such as brightness information, size information, shape information, color information, size information, etc. may include. The storage unit may store various data or information generated in the embodiment.

The apparatus 200 for detecting defects in an object according to an embodiment may include a labeling unit (not shown) that labels each of the detected or determined substances to distinguish them. Labeling can be done with colors, text, pictures, shapes, highlights, etc.

The apparatus 200 for detecting defects in an object according to an embodiment may include a calculation unit (or calculation unit, not shown) that calculates the size or area.

[First Example]

Figure 3 is a flowchart explaining a method for detecting defects in an object according to the first embodiment.

2 and 3, image data is acquired by the image acquisition unit 210 (S310), a plurality of features are extracted by the feature extraction unit 220 (S320), and the material determination unit (S320) At least one or more materials may be determined by 230 (S330), and defects for each of the determined one or more materials may be determined by the material determination unit 240 (S340).

Hereinafter, a method of determining at least one material using a plurality of characteristics will be described with reference to FIGS. 4 to 8.

Figure 4 shows detection of substances according to brightness.

As shown in FIG. 4, voids, resin, and filler can be determined using brightness, which is one of a plurality of characteristics.

Image data may include various substances classified by size, shape, color, etc. The brightness of these materials may also be different. The number of substances included in the image data according to their brightness can be displayed as shown in FIG. 4 . The unit of brightness may be gray-scale, but there is no limitation thereto.

The material corresponding to the 0 gradation and the A gradation may be a void, the material between the A and B gradation may be a resin, and the material corresponding to the B gradation or higher may be a filler.

Therefore, among the substances extracted from image data, those distributed between 0 and A gradations are determined as voids, those distributed between A and B gradations are determined as resin, and those distributed above B gradations are determined as resins. It can be determined by filler.

Unlike shown in FIG. 4, other materials may be used instead of voids, resin, and filler.

Figures 5a to 5f show detection of resin.

FIG. 5A is the original image data acquired by the image acquisition unit 210 shown in FIG. 2, and FIGS. 5B to 5F show that the area where resin is detected varies depending on different reference values.

Figure 5b shows the distribution of the resin 311 detected when the reference value of 5 pixels is exceeded, and only the resin 311 larger than the total area of 5 pixels can be detected. Figure 5c shows the distribution of the resin 311 detected when the reference value of 15 pixels is exceeded, and only the resin 311 larger than the total area of 15 pixels can be detected. Figure 5d shows the distribution of the resin 311 detected when it exceeds the reference value of 25 pixels, and only the resin 311 larger than the total area of 25 pixels can be detected. Figure 5e shows the distribution of the resin 311 detected when the reference value of 35 pixels is exceeded, and only the resin 311 larger than the total area of 35 pixels can be detected. Figure 5f shows the distribution of the resin 311 detected when it exceeds the reference value of 50 pixels, and only the resin 311 larger than the total area of 50 pixels can be detected.

As shown in FIGS. 5B to 5F, the degree of distribution of the detected resin 311 may vary depending on the size of the reference value. If the reference value is low or large, the amount of resin 311 detected may be small or large, and the accuracy of detecting the resin 311 may decrease, so setting an optimal reference value is required. For example, the optimal reference value may be 25 pixels, but there is no limitation thereto.

Meanwhile, the resin 311 may be detected using at least one of a plurality of features extracted from image data.

For example, resin 311 can be detected using brightness. For example, resin 311 can be detected using brightness and size. For example, resin 311 can be detected using brightness, size, and shape. For example, resin 311 can be detected using brightness, size, shape, and color.

Figures 6a and 6b show detection of filler.

A plurality of features may be extracted from image data (FIG. 6A), and a plurality of fillers 312 may be detected using at least one of the extracted features (FIG. 6B).

For example, filler 312 can be detected using brightness. For example, filler 312 can be detected using brightness and size. For example, filler 312 can be detected using brightness, size, and shape. For example, filler 312 can be detected using brightness, size, shape, and color.

Figures 7a and 7b show detection of voids.

By adjusting the sharpness in the image data (FIG. 7A), voids 314 can be detected (FIG. 7B).

In the image data shown in FIG. 7A, both the resin and the void 314 appear black, making it difficult to distinguish them from each other. Accordingly, the clarity may be adjusted to distinguish the void 314 from resin or other members. Even though both the void 314 and the resin appear black, the void 314 can be distinguished from the resin by increasing clarity.

As shown in FIG. 7B, by adjusting the clarity, the void 314 can be detected separately from the resin 311.

Meanwhile, as previously described, substances can be detected according to color.

Figure 8 shows detection of substances according to color.

As shown in FIG. 8, various materials can be distinguished by white and black in image data. Various substances can be detected through different colors including white and black.

Figure 8 shows a prepreg, which may include epoxy resin 313 and glass fabric 315.

For example, the epoxy resin 313 may be displayed in black, and the glass fabric 315 may be displayed in white.

Materials smaller than the size of the glass fabric 315 may be fillers. Both the glass fabric 315 and the filler may be displayed in white. In this case, if the sizes of the glass fabric 315 and the filler are known, the material corresponding to the corresponding size among the materials displayed in white can be detected as the glass fabric 315 or the filler.

When fillers of various sizes exist and the sizes of each of the various fillers are known, each of the fillers can be detected separately.

Meanwhile, as shown in FIG. 3, when at least one material is determined (S330), defectiveness of at least one material may be determined (S330).

Hereinafter, a method for determining defects in at least one material will be described with reference to FIGS. 9A to 12.

At least one material may include a void 314, a resin 311, and a filler 312, as shown in FIGS. 5 to 8.

As described above, at least one material, that is, the void 314, the resin 311 or the filler 312, can be determined using a plurality of features extracted from the image data.

Accordingly, the embodiment can easily determine what substance is contained in the object through analysis of image data. In addition, in the embodiment, it is possible to quickly determine whether the determined material is defective because it does not satisfy the preset conditions.

In order for an object to produce a product through a post-process, at least one substance contained in the object must satisfy previously established conditions. If at least one substance does not satisfy the preset conditions, not only is the substance determined to be defective, but the object may also be discarded as defective due to the defect in the substance.

In the embodiment, even if the type of substance contained in the object is not known, the type of substance contained in the object, that is, what substances are contained, is determined based on the image data acquired from the object, and the identified substance is determined under preset conditions. You can easily determine whether the material is defective by checking whether it satisfies the requirements.

A defect in a material can be determined based on at least one of area ratio, size, and density.

<Method for determining void defects using area ratio>

Figure 9a shows voids distributed in the AOI image. Figure 9b is a graph showing defect determination according to the area ratio of voids.

As shown in FIG. 9A, voids 314 may be distributed in the Automatic Optical Inspection (AOI) image 400. AOI image 400 may be an image acquired by an automatic optical inspection device. The AOI image 400 is an area for defect detection and may be the image data itself, or one or two or more blocks among a plurality of blocks separated from the image data.

As shown in FIG. 9B, the preset condition for determining defectiveness according to the area ratio of the void 314 may be, for example, 1%, but this is not limited. If the area ratio of the void 314 exceeds 1%, it may be judged as defective. The area ratio of the void 314 may be the ratio of the area occupied by the sum of the areas of each of the voids 314 distributed in the AOI image 400 to the area of the AOI image 400.

If the preset condition is 1%, defects are managed very strictly, and the preset condition may be managed at 10%. In this case, if the area ratio of the void 314 is 10% or more, the void 314 may be determined to be defective. That is, if the area ratio of the void 314 is 10% or more, poor conduction or moisture absorption may increase, and the possibility of metal protrusion (or protrusion) as a circuit pattern (or signal pattern) may increase. The preset conditions may be set differently depending on the size of the dielectric constant. For example, when the first dielectric constant is 2.0 to 3.5, the preset condition may be 5%, but this is not limited. For example, the second dielectric constant may be 3.5 to 6, and the third dielectric constant may be 6 to 10.

<Method for determining void defects using size>

Figure 10 shows how to determine a defect using the ferret length of the void.

As shown in FIG. 10, a void 314 may be included in the AOI image 400. The Feret length (or diameter) of the void 314 may be defined. Pallet length may be a measurement of the size of an object along a specified direction. Typically, it can be defined as the distance between two parallel planes that limit an object perpendicular to that direction. These measurements can be used for particle size analysis, such as microscopy, applied to projections of three-dimensional (3D) objects on a 2D plane. In this case, the ferret length (L1) can be defined as the distance between two parallel tangent lines.

For example, a preset condition for determining whether the ferret length L1 of the void 314 is defective may be 1㎛. In this case, if the ferret length L1 of the void 314 exceeds 1㎛, the void 314 may be determined to be defective.

Meanwhile, as shown in FIGS. 9A and 9B, the area ratio of the void 314 satisfies a preset condition, so the void 314 may be normal. However, if the ferret length of one of these voids 314 exceeds the preset condition of 1 μm as shown in FIG. 10, the void 314 may ultimately be determined to be defective.

<How to determine resin defects using size>

Figure 11 shows how to determine defects using the ferret length of the resin.

As shown in FIG. 11, a plurality of resins 311 may be distributed in the AOI image 400. The largest ferret length (L2) may be selected among the plurality of resins 311.

For example, the preset condition for determining whether the ferret length (L2) of the resin 311 is defective may be 20% of the long axis length (L_long) in the long axis * short axis of the AOI image 400. In this case, if the ferret length (L2) of the resin 311 exceeds 20% of the major axis length (L_long) of the AOI image 400, the resin 311 may be determined to be defective. L_short may mean shortened length.

<Method for determining agglomeration defects in resin using density>

Figure 12a shows the agglomeration area of distributed resin in the AOI image. Figure 12b is a graph showing defective judgment according to the area ratio of the resin agglomeration area.

As shown in FIG. 12A, agglomerated areas 410 where resin is agglomerated may be distributed in the AOI image 400. For example, in the agglomeration area 410, at least two or more resins may be agglomerated. Since the resins stick together or clump together rather than being distributed apart from each other, it is difficult to apply defect determination using the void area ratio (FIGS. 9A and 9B). Accordingly, in the case of resin, it is desirable to determine defects by considering density, that is, the degree of agglomeration.

For example, when agglomerated areas 410 where resin is agglomerated are distributed in the AOI image 400, defective resin may be determined depending on whether preset conditions are satisfied. Here, the preset condition may be 20% of the AOI area. In this case, as shown in Figure 12b, if the area ratio of the resin agglomeration area 410 exceeds 20%, the resin may be determined to be defective.

The resin agglomeration area 410 may be selected from areas where filler or glass fabric does not exist to increase the accuracy of defect determination.

Meanwhile, in order to select the agglomeration area 410 of the resin, a specific size must be defined so that the agglomeration area 410 of the resin can be selected below the specific size. For example, the resin agglomeration area 410 may be selected to be less than 0.015 to 1.535% of the AOI image 400 having a specific size of 768*666.

On the other hand, when a metal such as a circuit pattern is present in the object, when histogram analysis is performed on the AOI image 400 acquired for these metals, the brightness value corresponding to white of at least 255 levels for the metal will be expressed. You can. In this case, the brightness value for the metal corresponding to white has no value as information and needs to be removed.

[Second Embodiment]

Figure 13 is a flowchart explaining a method for detecting defects in an object according to a second embodiment.

As shown in Figures 2 and 13, image data is acquired by the image acquisition unit 210 (S310), histogram extreme value correction is performed on the acquired image data (S350), and the feature extraction unit (S350) A plurality of features are extracted by 220 (S320), and at least one material can be determined by the material determination unit 230 (S330).

S310 may be performed by an image acquisition unit or a feature extraction unit. S310 may be performed by a third device or unit other than the image acquisition unit or feature extraction unit.

When histogram analysis is performed on image data acquired from an object with a circuit pattern, as shown in FIG. 14, there may be a plurality of brightness values of D or more. The number corresponding to each brightness value above D reflects the reflectivity caused by the metal component of the circuit pattern, and areas above the brightness value above D have no value as information and need to be removed. Here, D may be, for example, a brightness value of 250 levels, but this is not limited.

According to an embodiment, when the object includes a circuit by a histogram extreme value correction unit (not shown), histogram extreme value correction may be performed based on the acquired image data. For example, as shown in FIG. 14, after image data corresponding to an area with a brightness value of D or higher is removed, the AOI image may be reset based on the remaining data. In this case, the AOI image may not include information about the circuit pattern.

Meanwhile, the histogram extreme value correction unit may be included in the image acquisition unit or the feature extraction unit, but this is not limited.

The above detailed description should not be construed as restrictive in any respect and should be considered illustrative. The scope of the embodiments should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the embodiments are included in the scope of the embodiments.

110: prepreg
120: Resin layer
130: filler
141 to 143: Circuit pattern
151, 152: Via
200, 201: Device for detecting defects in an object
210: Image acquisition unit
220: feature extraction unit
230: material crystal unit
240: Defect determination unit
311: Resin
312: Filler
313: Epoxy resin
314: void
315: glass fabric
400: AOI image
410: Agglomeration area

Claims (18)

Obtaining image data for a specific area of the object;
determining at least one substance based on the acquired image data; and
Including, determining whether the determined material is defective.
Method for detecting defects in an object. According to paragraph 1,
The determined materials include voids, resins and fillers,
Method for detecting defects in an object. According to paragraph 2,
The defect of the determined material is determined based on at least one of area ratio, size, and density,
Method for detecting defects in an object. According to paragraph 3,
The step of determining whether the determined material is defective is,
Including, determining whether the void is defective based on the area ratio of the void.
Method for detecting defects in an object. According to paragraph 3,
The step of determining whether the determined material is defective is,
Including; determining whether the void is defective based on the ferret length of the void.
Method for detecting defects in an object. According to paragraph 3,
The step of determining whether the determined material is defective is,
Including; determining whether the resin is defective based on the length of the resin.
Method for detecting defects in an object. According to paragraph 3,
The step of determining whether the determined material is defective is,
Including, determining whether the resin is defective based on the area ratio of the agglomeration area of the resin determined as the density.
Method for detecting defects in an object. In clause 7,
The agglomeration area of the resin is an area less than 0.015 to 1.535% of the AOI size,
Method for detecting defects in an object. According to paragraph 1,
When the object includes a circuit, performing histogram extreme value correction based on the acquired image data,
Method for detecting defects in an object. An image acquisition unit that acquires image data for a specific area of the object;
a material determination unit that determines at least one material based on the acquired image data; and
Including a defect determination unit that determines whether the determined material is defective.
A device for detecting defects in an object. According to clause 10,
The determined materials include voids, resins and fillers,
A device for detecting defects in an object. According to clause 11,
The defect of the determined material is determined based on at least one of area ratio, size, and density,
A device for detecting defects in an object. According to clause 12,
The defect determination unit,
Determining whether the void is defective based on the area ratio of the void,
A device for detecting defects in an object. According to clause 12,
The defect determination unit,
Determining whether the void is defective based on the ferret length of the void,
A device for detecting defects in an object. According to clause 12,
The defect determination unit,
Determining defects in the resin based on the length of the resin.
A device for detecting defects in an object. According to clause 12,
The defect determination unit,
Determining whether the resin is defective based on the area ratio of the agglomeration area of the resin determined as the density,
A device for detecting defects in an object. According to clause 16,
The agglomeration area of the resin is an area of 0.015 to 1.535% or less of the AOI size,
A device for detecting defects in an object. According to clause 10,
When the object includes a circuit, a histogram extreme value correction unit that performs histogram extreme value correction based on the acquired image data,
A device for detecting defects in an object.

Images