Apparatus and method for detecting defects in an object

Abstract

The method for detecting defects in an object includes acquiring image data of a specific region of the object, determining at least one material based on the acquired image data, and discriminating defects in the determined material. This allows various materials contained in the object to be easily identified through the image data acquired from the object. Furthermore, since materials are identified based on the image acquired from the object, the structure is simple and costs can be reduced.

Description

[Technical Field]

[0001] The embodiments relate to an apparatus and a method for detecting defects in an object. [Background technology]

[0002] Manufacturers who produce products based on objects such as raw materials want to know the types and characteristics of substances contained in the raw materials, but because raw material suppliers do not provide this information for reasons such as information security, they are unable to find out the types and characteristics of substances contained in the raw materials.

[0003] If a manufacturer knows the types of substances contained in raw materials and their properties, they can understand the causality of how these substances change or are involved in the manufacturing process of a product, and minimize defects that may occur during product production based on that causality. Also, if a manufacturer knows the types of substances contained in raw materials and their properties, they can provide feedback to raw material suppliers on changing the types of substances or adjusting the properties of materials to minimize defects that may occur during product production, and in the future they can receive better quality raw materials from raw material suppliers and manufacture better quality products.

[0004] Therefore, there is a strong demand for a technology that can easily grasp the types and characteristics of materials contained in a target object. Summary of the Invention [Problem to be solved by the invention]

[0005] The embodiments are directed to solving the above-mentioned problems and other problems.

[0006] Another object of the embodiments is to provide an apparatus and method for detecting defects in an object, which can easily identify substances contained in the object.

[0007] Another object of the present invention is to provide an apparatus and method for detecting defects in an object that are simple in structure and can reduce costs by detecting material based on an image acquired from the object.

[0008] Still another object of the present invention is to provide an apparatus and method for detecting defects in an object, which can easily identify defects in the grasped substance.

[0009] The technical problems of the embodiments are not limited to those described in this section, but include those that can be understood from the description of the invention. [Means for solving the problem]

[0010] To achieve the above or another object, according to one aspect of the embodiment, a method for detecting defects in an object includes the steps of acquiring image data for a specific region of the object, determining at least one material based on the acquired image data, and determining defects in the determined materials.

[0011] The determined materials may include voids, resins, and fillers.

[0012] The determined material defects may be distinguished based on at least one of area ratio, size, and density.

[0013] The step of determining whether the determined material is defective may include determining whether the void is defective based on an area ratio of the void.

[0014] The step of determining whether the determined material is defective may include determining whether the void is defective based on a Feret length of the void.

[0015] The step of determining whether the determined material is defective may include determining whether the resin is defective based on a ferret length of the resin.

[0016] The step of determining whether the determined material is defective may include determining whether the resin is defective based on an area ratio of the determined resin lump region to the density.

[0017] The resin lump region may be a region that is 0.015 to less than 1.535% of the AOI size.

[0018] The method for detecting defects in an object may include performing histogram extremum correction based on the acquired image data when the object includes a circuit.

[0019] To achieve the above or another object, according to another aspect of the embodiment, a defect detection device for an object includes an image acquisition unit that acquires image data for a specific region 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 defects in the determined materials.

[0020] The determined materials may include voids, resins, and fillers.

[0021] The determined material defects may be distinguished based on at least one of area ratio, size, and density.

[0022] The defect determining section can determine whether the void is defective based on the area ratio of the void.

[0023] The defect determining unit can determine whether the void is defective based on the ferret length of the void.

[0024] The defect determining unit can determine whether the resin is defective based on the ferret length of the resin.

[0025] The defect determining section can determine whether the resin is defective based on the area ratio of the resin lump region determined as the density.

[0026] The resin lump region may be a region of 0.015 to 1.535% or less of the AOI size.

[0027] The object defect detection device 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. [Effects of the Invention]

[0028] The effects of the object defect detection device and object defect detection method according to the embodiment will be described below.

[0029] According to at least one of the embodiments, it is possible to easily grasp various materials contained in the object through image data acquired from the object.

[0030] According to at least one of the embodiments, since a material is identified based on an image acquired from a target object, the structure is simple and the cost can be reduced.

[0031] According to at least one of the embodiments, it is possible to easily determine whether each of the at least one or more determined (identified) substances is defective or not, thereby making it possible to not only determine the type of substance contained in the object but also determine whether the type is defective or not, thereby enabling three-dimensional inspection of the object.

[0032] Further scope of applicability of the embodiments will become apparent from the following detailed description. However, it should be understood that the detailed description and specific embodiments, such as the preferred embodiment, are merely illustrative, as various changes and modifications within the spirit and scope of the embodiments will be apparent to those skilled in the art. [Brief explanation of the drawings]

[0033] [Figure 1] FIG. 1 is a cross-sectional view illustrating a circuit board according to an embodiment. [Figure 2]FIG. 2 is a block diagram illustrating an apparatus for detecting defects in an object according to an embodiment. [Figure 3] FIG. 3 is a flowchart illustrating a method for detecting defects in an object according to the first embodiment. [Figure 4] FIG. 4 illustrates the detection of materials depending on brightness. [Figure 5a] FIG. 5a illustrates the detection of the resin. [Figure 5b] FIG. 5b illustrates the detection of the resin. [Figure 5c] FIG. 5c illustrates the detection of the resin. [Figure 5d] FIG. 5d illustrates the detection of the resin. [Figure 5e] FIG. 5e illustrates the detection of the resin. [Figure 5f] FIG. 5f illustrates the detection of the resin. [Figure 6a] FIG. 6a illustrates how fillers are detected. [Figure 6b] FIG. 6b illustrates how fillers are detected. [Figure 7a] FIG. 7a illustrates how voids are detected. [Figure 7b] FIG. 7b illustrates how voids are detected. [Figure 8] FIG. 8 illustrates detecting substances according to color. [Figure 9a] Figure 9a shows the voids distributed in the AOI image. [Figure 9b] FIG. 9b is a graph showing the determination of defects based on the void area ratio. [Figure 10] FIG. 10 shows how defects are identified using the ferret length of voids. [Figure 11] FIG. 11 shows how defects are identified using the ferret length of the resin. [Figure 12a] Figure 12a shows the resin agglomerate areas distributed in the AOI image. [Figure 12b]FIG. 12b is a graph showing the determination of defects based on the area ratio of the resin lump region. [Figure 13] FIG. 13 is a flowchart illustrating a method for detecting defects in an object according to the second embodiment. [Figure 14] FIG. 14 shows histogram extreme value correction. DETAILED DESCRIPTION OF THE INVENTION

[0034] Hereinafter, the embodiments disclosed herein will be described in detail with reference to the accompanying drawings. Regardless of the reference numerals, identical or similar components will be designated by the same reference numerals, and redundant description thereof will be omitted. The suffixes "module" and "section" used in the following description are used interchangeably to facilitate the preparation of the specification and do not have any distinguishing meaning or function. The accompanying drawings are provided solely for the purpose of facilitating understanding of the embodiments disclosed herein, and are not intended to limit the technical ideas disclosed herein. Furthermore, when an element such as a layer, region, or substrate is referred to as being "on" another element, this includes whether it is directly on the other element or whether other intermediate elements may exist therebetween.

[0035] The embodiments may provide a method and apparatus that can easily identify the type of substance contained in an object such as a raw material.

[0036] In the following description, a circuit board is typically used as a raw material, but the embodiment may include a raw material containing at least one substance. The embodiment may also include a component containing at least one substance other than the raw material.

[0037] The substances described below may be tangible, such as resins, fillers, epoxy resins, or glass fabrics, or formless, such as voids. The terms substance and object are sometimes used interchangeably.

[0038] FIG. 1 is a cross-sectional view illustrating a circuit board according to an embodiment.

[0039] Referring to FIG. 1, the circuit board according to the embodiment may include a plurality of prepregs 110, a resin layer 120 between the prepregs 110, and a plurality of vias 151, 152.

[0040] The prepreg 110 may be formed by impregnating a fiber layer in the form of a woven sheet, such as a glass fabric 112 woven with glass fiber yarn, with an epoxy resin 111 and then subjecting the layer to heat compression. However, the embodiment is not limited thereto. That is, the prepreg 110 may also include a fiber layer in the form of a woven sheet woven with carbon fiber yarn.

[0041] After the resin layer 120 is placed between the prepregs 110, the substrate is completed by applying pressure using a thermocompression method using a press. Then, via holes are formed by drilling the prepregs 110, and the via holes are filled with or plated with a conductive material to form vias 151 and 152. The conductive material may be any one selected from Cu, Ag, Sn, Au, Ni, and Pd. The via holes can be formed using any one of mechanical, laser, and chemical processing methods.

[0042] The vias 151 and 152 are electrically connected to the circuit patterns 141 to 143 provided on the upper and / or lower sides of the prepreg 110 .

[0043] Meanwhile, the undescribed reference numeral 113 is a void, which is an empty space where no substance exists. Also, the undescribed reference numeral 130 is a filler. The filler 130 may include various different types of fillers.

[0044] FIG. 2 is a block diagram illustrating an apparatus for detecting defects in an object according to an embodiment.

[0045] In the following description, for convenience of explanation, the substrate illustrated in FIG. 1 is limited as the subject matter, but the embodiments can be similarly applied to other members other than the substrate illustrated in FIG. 1.

[0046] 1 and 2, the object defect detection device 200 according to the embodiment may include an image acquisition unit 210, a feature extraction unit 220, a material determination unit 230, and a defect determination unit 240.

[0047] The image acquiring unit 210 can acquire image data for a specific region of the object.

[0048] The image data may be acquired by a destructive testing method. 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 cross section. For example, the image data may be acquired by a microscope, SEM, TEM, etc., but is not limited thereto.

[0049] Meanwhile, the embodiment may also acquire image data for a specific area to be photographed using a non-destructive inspection method.

[0050] The feature extraction unit 220 can extract a plurality of features included in the image data, such as brightness, size, shape, color, and sharpness.

[0051] Image data includes various materials that can be distinguished from one another based on brightness, boundary, color, definition, etc. That is, these materials can be distinguished from one another based on brightness, boundary, color, definition, etc. The substrate illustrated in FIG. 1 may refer to various different materials. That is, the substrate is formed by a composition of various materials. These materials may be resin, glass fabric, filter, etc. Voids can also be classified as a type of material as they are included in the substrate.

[0052] The material determination unit 230 may determine at least one material based on at least one of the 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.

[0053] Since the substrate illustrated in FIG. 1 includes resin, at least one or more fillers, glass fabric, and voids as materials, at least one or more of the resin, at least one or more fillers, glass fabric, and voids may be determined based on at least one or more of brightness, size, shape, and color.

[0054] For example, the material determining unit 230 may determine at least one material based on brightness.

[0055] For example, the material determining unit 230 may determine at least one material based on brightness and size.

[0056] For example, the material determining unit 230 may determine at least one material based on brightness, size, and shape.

[0057] For example, the material determiner 230 may determine at least one material based on brightness, size, shape, and color.

[0058] Meanwhile, the material determination unit 230 may determine voids based on clarity. As shown in FIG. 7A, when viewing image data in color, both resin and voids appear black, making them indistinguishable from each other and resulting in errors in determining voids and resin. Therefore, by considering clarity as a characteristic, resin and voids may be distinguished based on the magnitude of clarity. By increasing clarity, the boundaries of voids may be clearly visible. In this case, taking into account that voids are relatively much smaller than resin, clarity may be increased to determine that a material with an outline that is much smaller than resin is a void.

[0059] Meanwhile, the defect determination unit 240 can determine whether or not the at least one or more determined substances are defective. The quality of each of the at least one or more substances may be determined based on whether or not the substance satisfies a predetermined condition.

[0060] Meanwhile, the object defect detection device 200 according to the embodiment may include a storage unit (not shown) that stores various information of image data and materials, i.e., brightness information, size information, shape information, color information, size information, etc. The storage unit stores various data and information generated in the embodiment.

[0061] The object defect detection device 200 according to the embodiment may include a labeling unit (not shown) for labeling each of the detected or determined substances. The labeling may be processed using color, text, pictures, shapes, highlights, etc.

[0062] The object defect detection device 200 according to the embodiment may include a calculation unit (or an operation unit, not shown) that calculates the size or area.

[0063] [First Example]

[0064] FIG. 3 is a flowchart illustrating a method for detecting defects in an object according to the first embodiment.

[0065] 2 and 3, image data may be acquired by the image acquisition unit 210 (S310), a plurality of features may be extracted by the feature extraction unit 220 (S320), and at least one material may be determined by the material determination unit 230 (S330). A defect may be determined for each of the at least one material determined by the material determination unit 240 (S340).

[0066] Hereinafter, a method for determining at least one substance using a plurality of features will be described with reference to FIGS.

[0067] FIG. 4 illustrates the detection of materials depending on brightness.

[0068] As shown in FIG. 4, brightness is one of several features used to determine voids, resin, and filler.

[0069] Image data may include various materials that are differentiated by size, shape, color, etc. The brightness of these materials may also be different. The number of materials included in the image data according to their brightness may be illustrated as shown in FIG. 4. The unit of brightness may be gray-scale, but is not limited thereto.

[0070] A material between 0 gradation and A gradation is a void, a material between A gradation and B gradation is a resin, and a material above B gradation may be a filler.

[0071] Therefore, among the materials extracted from the video data, materials distributed between 0 gradation and A gradation may be determined as voids, materials distributed between A gradation and B gradation may be determined as resins, and materials distributed above B gradation may be determined as fillers.

[0072] Unlike what is shown in FIG. 4, other materials may be determined instead of the voids, resin, and filler.

[0073] 5a to 5f illustrate how the resin is detected.

[0074] FIG. 5a shows the original image data acquired from the image acquisition unit 210 shown in FIG. 2, and FIGS. 5b to 5f show that the area where the resin is detected changes according to different reference values.

[0075] FIG. 5b shows a distribution of resin 311 detected when the reference value of 5 pixels is exceeded, where only resin 311 with an area larger than the entire area of ​​5 pixels may be detected. FIG. 5c shows a distribution of resin 311 detected when the reference value of 15 pixels is exceeded, where only resin 311 with an area larger than the entire area of ​​15 pixels may be detected. FIG. 5d shows a distribution of resin 311 detected when the reference value of 25 pixels is exceeded, where only resin 311 with an area larger than the entire area of ​​25 pixels may be detected. FIG. 5e shows a distribution of resin 311 detected when the reference value of 35 pixels is exceeded, where only resin 311 with an area larger than the entire area of ​​35 pixels may be detected. FIG. 5f shows a distribution of resin 311 detected when the reference value of 50 pixels is exceeded, where only resin 311 with an area larger than the entire area of ​​50 pixels may be detected.

[0076] As shown in Figures 5b to 5f, the distribution of detected resin 311 varies depending on the magnitude of the reference value. If the reference value is low or high, the amount of detected resin 311 will be small or large, which may reduce the accuracy of resin 311 detection, so an optimal reference value setting is required. For example, the optimal reference value may be 25 pixels, but is not limited to this.

[0077] On the other hand, the resin 311 may be detected by using at least one of a plurality of features extracted from the video data.

[0078] For example, the resin 311 may be detected using brightness. For example, the resin 311 may be detected using brightness and size. For example, the resin 311 may be detected using brightness, size, and shape. For example, the resin 311 may be detected using brightness, size, shape, and color.

[0079] 6a and 6b illustrate how fillers are detected.

[0080] A plurality of features may be extracted from the video data (FIG. 6a), and at least one of the extracted features may be used to detect a plurality of fillers 312 (FIG. 6b).

[0081] For example, the filler 312 may be detected using brightness. For example, the filler 312 may be detected using brightness and size. For example, the filler 312 may be detected using brightness, size, and shape. For example, the filler 312 may be detected using brightness, size, shape, and color.

[0082] 7a and 7b illustrate how voids are detected.

[0083] By adjusting the sharpness of the image data (FIG. 7a), a void 314 is detected (FIG. 7b).

[0084] In the image data shown in Figure 7a, the resin and voids 314 both appear black and are difficult to distinguish from each other. Therefore, the clarity is adjusted to distinguish the voids 314 from the resin or other components. Even though the voids 314 and resin both appear black, increasing the clarity allows the voids 314 to be distinguished from the resin.

[0085] As shown in FIG. 7b, by adjusting the visibility, the void 314 can be detected as being distinguished from the resin 311.

[0086] On the other hand, as mentioned above, substances are detected according to their color.

[0087] FIG. 8 illustrates detecting substances according to color.

[0088] As shown in Figure 8, various materials are classified into white and black in the image data, and various materials are detected according to the difference in color including white and black.

[0089] FIG. 8 illustrates a prepreg that can include epoxy resin 313 and glass fabric 315 .

[0090] For example, the epoxy resin 313 may be displayed in black and the glass fabric 315 may be displayed in white.

[0091] A material smaller than the size of the glass fabric 315 may be a filler. 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 displayed in white that corresponds to the size is detected as the glass fabric 315 or the filler.

[0092] When fillers of various sizes are present, if the sizes of the various fillers are known, each of the fillers can be detected separately.

[0093] Meanwhile, as shown in FIG. 3, when at least one material is determined (S330), defects of at least one material may be determined (S330).

[0094] Hereinafter, a method for determining defects in at least one or more materials will be described with reference to FIGS. 9a to 12. FIG.

[0095] The at least one or more substances may include voids 314, resin 311, and filler 312, as illustrated in FIGS.

[0096] As described above, at least one substance, namely, void 314, is determined to be resin 311 or filler 312 using a plurality of features extracted from the image data.

[0097] Therefore, the embodiment can easily determine what substances are contained in the target object through analysis of the image data. Furthermore, the embodiment can easily determine whether the determined substances are defective because they do not satisfy the preset conditions.

[0098] In order for an object to be used in a subsequent process to produce a product, at least one substance contained in the object must satisfy a predetermined condition. If at least one substance does not satisfy the predetermined condition, not only will the substance be judged as defective, but the object will also be discarded as defective due to the defective substance.

[0099] In the embodiment, even if the type of substance contained in the object is unknown, the type of substance contained in the object, i.e., what kind of substance is contained, can be determined based on the image data acquired from the object, and the quality of the substance can be easily determined based on whether the determined substance satisfies the predetermined conditions.

[0100] Material defects may be determined based on at least one of area ratio, size, and density.

[0101] <Method for determining defects of voids using area ratio>

[0102] Figure 9a shows voids distributed in an AOI image, and Figure 9b is a graph showing the failure discrimination based on the void area ratio.

[0103] 9A, voids 314 are distributed in an AOI (Automatic Optical Inspection) image 400. The AOI image 400 may be an image acquired by an automatic optical inspection device. The AOI image 400 may be the image data itself, or one or more blocks among a plurality of blocks segmented from the image data, as an area where defects are to be detected.

[0104] 9b, the preset condition for determining whether a void 314 is defective based on the area ratio of the voids 314 may be, for example, 1%, but is not limited thereto. If the area ratio of the voids 314 exceeds 1%, the void 314 may be determined to be defective. The area ratio of the voids 314 may be the ratio of the area of ​​the AOI image 400 to the area of ​​the AOI image 400, which is the sum of the areas of the voids 314 distributed in the AOI image 400.

[0105] When the preset condition is 1%, defects are controlled very strictly, and the preset condition may be controlled to 10%. In such a case, if the area ratio of voids 314 is 10% or more, the voids 314 may be determined to be defective. That is, if the area ratio of voids 314 is 10% or more, electrical conductivity failure and moisture absorption failure increase, and the possibility of metal protrusions (or projections) as circuit patterns (or signal patterns) increases. The preset condition may be set differently depending on the magnitude of the dielectric constant. For example, when the first dielectric constant is 2.0 to 3.5, the preset condition may be 5%, but is not limited thereto. For example, the second dielectric constant may be 3.5 to 6, and the third dielectric constant may be 6 to 10.

[0106] <Method for determining void defects using size>

[0107] FIG. 10 shows how defects are identified using the ferret length of voids.

[0108] As shown in FIG. 10, the AOI image 400 may contain a void 314. The ferret length (or diameter) of the void 314 can be defined. The ferret 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 bound the object perpendicular to that direction. Such measurements can be used for particle size analysis, such as microscopy, which applies to the projection of a three-dimensional (3D) object on a 2D plane. In such cases, the ferret length L1 can be defined as the distance between two parallel tangents.

[0109] For example, the preset condition for determining whether the ferret length L1 of the void 314 is defective may be 1 μm. In this case, if the ferret length L1 of the void 314 exceeds 1 μm, the void 314 may be determined to be defective.

[0110] 9A and 9B, the area ratio of the voids 314 may satisfy a preset condition, and the voids 314 may be normal. However, if the ferret length of one of the voids 314 exceeds the preset condition of 1 μm, as shown in FIG. 10, the void 314 may ultimately be determined to be defective.

[0111] <Method for identifying resin defects using size>

[0112] FIG. 11 shows how defects are identified using the ferret length of the resin.

[0113] 11, a plurality of resins 311 are distributed in the AOI image 400. The largest ferret length L2 of the plurality of resins 311 may be selected.

[0114] For example, a 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 x short axis of the AOI image 400. In this case, if the ferret length L2 of the resin 311 exceeds 20% of the long axis length (L_long) of the AOI image 400, the resin 311 may be determined to be defective. L_short may refer to the short axis length.

[0115] <Method for identifying defective resin lumps using density>

[0116] Figure 12a shows resin lump regions distributed in an AOI image, and Figure 12b is a graph showing defective discrimination based on the area ratio of resin lump regions.

[0117] As shown in Figure 12a, clump regions 410 where resin has aggregated are distributed in the AOI image 400. For example, clump region 410 is composed of at least two or more resin particles. Because resin particles tend to clump together rather than being spaced apart, it is difficult to apply defect detection based on the void area ratio (Figures 9a and 9b). For this reason, it is preferable to determine defects in resin by considering its density, i.e., the degree of clumping.

[0118] For example, if resin lump regions 410 are distributed in the AOI image 400, the resin may be determined to be defective depending on whether a preset condition is satisfied. Here, the preset condition may be a ratio of 20% to the AOI area. In this case, as shown in FIG. 12b, if the area ratio of the resin lump regions 410 exceeds 20%, the resin may be determined to be defective.

[0119] The resin lump region 410 is selected in an area where no filler or glass fabric exists in order to increase the accuracy of defect detection.

[0120] On the other hand, a specific size needs to be defined to select the resin lump region 410, and the resin lump region 410 may be selected within a size less than the specific size. For example, the specific size may be 0.015 to 1.535% of the AOI image 400 having a size of 768 x 666, and the resin lump region 410 may be selected within that size.

[0121] On the other hand, if the object contains metal such as a circuit pattern, when a histogram analysis is performed on the AOI image 400 acquired for the metal, the metal can be expressed as a brightness value corresponding to white at least equal to 255. In this case, the brightness value of the metal corresponding to white has no information value and needs to be removed.

[0122] [Second Example]

[0123] FIG. 13 is a flowchart illustrating a method for detecting defects in an object according to the second embodiment.

[0124] As shown in FIGS. 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), a plurality of features are extracted by the feature extraction unit 220 (S320), and at least one material is determined by the material determination unit 230 (S330).

[0125] S310 may be performed by the image capture unit or the feature extraction unit, or S310 may be performed by a third device or unit instead of the image capture unit or the feature extraction unit.

[0126] When a histogram analysis is performed on image data acquired from an object having a circuit pattern, a large number of counts may exist for each brightness value equal to or greater than D, as shown in Fig. 14. The counts corresponding to each brightness value equal to or greater than D reflect the reflectivity of the metal component of the circuit pattern, and areas with brightness values ​​equal to or greater than D need to be removed because they have no information value. Here, D may be, for example, a brightness value of 250, but is not limited thereto.

[0127] According to an embodiment, if the object includes a circuit, a histogram extreme value correction unit (not shown) performs histogram extreme value correction based on the acquired image data. For example, as shown in Fig. 14, image data corresponding to an area with a brightness value of D or more is removed, and then the AOI image is re-established based on the remaining data. In this case, the AOI image does not include information about the circuit pattern.

[0128] Meanwhile, the histogram extreme value correction unit may be included in the image acquisition unit or the feature extraction unit, but is not limited thereto.

[0129] The above detailed description should not be construed as limiting in all respects, but should be considered as illustrative. The scope of the embodiments should be determined by a reasonable analysis of the appended claims, and all modifications within the equivalent range of the embodiments are included in the scope of the embodiments.

Claims

1. acquiring image data for a specific region of the object; determining at least one material based on the acquired image data; and determining whether the determined material is defective.

2. The method for detecting defects in an object according to claim 1 , wherein the determined substances include voids, resins, and fillers.

3. The method for detecting defects in an object according to claim 2 , wherein the determined defects of the material are determined based on at least one of an area ratio, a size, and a density.

4. The step of determining whether the determined material is defective comprises: The method for detecting defects in an object according to claim 3 , further comprising the step of determining whether the voids are defective based on an area ratio of the voids.

5. The step of determining whether the determined material is defective comprises: The method for detecting defects in an object according to claim 3, further comprising the step of determining whether the void is defective based on a Feret length of the void.

6. The step of determining whether the determined material is defective comprises: The method for detecting defects in an object according to claim 3 , further comprising the step of determining whether the resin is defective based on a ferret length of the resin.

7. The step of determining whether the determined material is defective comprises:

4. The method for detecting defects in an object according to claim 3, further comprising the step of determining whether the resin is defective based on the area ratio of the resin lump region determined as the density.

8. 8. The method for detecting defects in an object according to claim 7, wherein the resin lump region is an area that is 0.015 to less than 1.535% of the AOI size.

9. 2. The method of claim 1, further comprising: performing histogram extremum correction based on the acquired image data when the object includes a circuit.

10. an image acquisition unit for acquiring image data for a specific region of the object; a material determination unit that determines at least one material based on the acquired image data; a defect determination unit that determines whether the determined substance is defective.

11. The apparatus for detecting defects in an object according to claim 10 , wherein the determined substances include voids, resins, and fillers.

12. The apparatus for detecting defects in an object according to claim 11 , wherein the determined defects in the material are determined based on at least one of an area ratio, a size, and a density.

13. The apparatus for detecting defects in an object according to claim 12 , wherein the defect determining unit determines whether the voids are defective based on an area ratio of the voids.

14. The apparatus for detecting defects in an object according to claim 12 , wherein the defect determination unit determines whether the void is defective based on a Feret length of the void.

15. The apparatus for detecting defects in an object according to claim 12 , wherein the defect determining unit determines whether the resin is defective based on a ferret length of the resin.

16. 13. The apparatus for detecting defects in an object according to claim 12, wherein the defect determining unit determines whether the resin is defective based on an area ratio of the resin lump region determined as the density.

17. 17. The apparatus for detecting defects in an object according to claim 16, wherein the resin lump region is an area of ​​0.015 to 1.535% or less of the AOI size.

18. The apparatus for detecting defects in an object according to claim 10 , further comprising a histogram extreme value correction unit for correcting histogram extreme values ​​based on the acquired image data when the object includes a circuit.

Images