US-12621973-B2 - Image processing device, mounting device, and image processing method

US12621973B2US 12621973 B2US12621973 B2US 12621973B2US-12621973-B2

Abstract

An image processing device for processing an image of a tape having provided therein multiple cavities for receiving a component as a supply target includes a feature amount acquisition section configured to extract brightnesses of pixels within a predetermined range which can include a bottom surface of the cavity and a component inside the cavity using the image and acquire a value indicating scatter of the extracted brightnesses as a feature amount, and a determination section configured to determine on existence or non-existence of a component inside the cavity based on a determination threshold determined based on the feature amount for each of a case in which a component exists inside the cavity and a case in which no component exists inside the cavity and the feature amount acquired from the image.

Inventors

Yuki INAURA

Assignees

FUJI CORPORATION

Dates

Publication Date: 20260505
Application Date: 20200408

Claims (7)

1 . An image processing device for processing an image of a tape that includes multiple cavities for receiving a component as a supply target, the image processing device comprising: a feature amount acquisition section configured to extract brightnesses of pixels within a predetermined range which can include a bottom surface of a cavity of the multiple cavities and the component inside the cavity using the image and acquire a value indicating scatter of the extracted brightnesses as a feature amount; and a determination section configured to determine on existence or non-existence of the component inside the cavity based on a determination threshold determined based on the feature amount for each of a case in which the component exists inside the cavity and a case in which no component exists inside the cavity and the feature amount acquired from the image.
2 . The image processing device according to claim 1 , wherein the predetermined range is defined into a range resulting by adding a predetermined margin to a size of the component about a supply position for use when the tape supplies the component inside the cavity, and wherein the feature amount acquisition section extracts brightnesses of the pixels from an image resulting by converting the predetermined range in the image into a specific size and averaging the predetermined range so converted.
3 . The image processing device according to claim 1 , further comprising: a tape type acquisition section configured to acquire a type of the tape as a processing target, wherein the determination section uses a threshold determined individually for types of the tape, the threshold corresponding to the type of the tape acquired by the tape type acquisition section.
4 . The image processing device according to claim 1 , further comprising: a component type acquisition section configured to acquire a type of the component as a supply target, wherein the determination section uses a threshold in determined individually for types of the component, the threshold corresponding to the type of the component acquired by the component type acquisition section.
5 . The image processing device according to claim 1 , wherein the feature amount acquisition section acquires multiple feature amounts including at least one of a brightness dispersion or a standard deviation as a value indicating scatter of brightnesses, and wherein the determination section determines on existence or non-existence of the component inside the cavity based on a threshold determined through machine learning using the multiple feature amounts as variables and the multiple feature amounts acquired from the image.
6 . A mounting device to which a feeder is attached and which is configured to mount a component as a supply target, comprising: the image processing device according to claim 1 ; and an imaging device configured to image the tape, wherein the feeder feeds the tape, and wherein the component is picked up from the cavity.
7 . An image processing method for processing an image of a tape that includes multiple cavities for receiving a component as a supply target, the image processing method comprising: (a) a step of extracting brightnesses of pixels within a predetermined range which can include a bottom surface of the cavity and the component inside a cavity of the multiple cavities using the image and acquiring a value indicating scatter of the extracted brightnesses as a feature amount; and (b) a step of determining on existence or non-existence of the component inside the cavity based on a determination threshold determined based on the feature amount for each of a case in which the component exists inside the cavity and a case in which no component exists inside the cavity and the feature amount acquired from the image.

Description

TECHNICAL FIELD The present description discloses an image processing device, a mounting device, and an image processing method. BACKGROUND ART Conventionally, there has been proposed a device for capturing an image of a tape in which multiple cavities for receiving a component are provided to thereby determine on existence or non-existence of a component inside a cavity based on a feature amount acquired from the image (for example, refer to Patent Literature 1). With this device, a mean value of brightnesses of pixels is calculated as a feature amount, and the mean value so calculated is compared with a threshold set based on the brightness of a body portion of a component or a threshold set based on the brightness of a cavity to thereby determine on existence or non-existence of component. As a result, a retry operation, which is triggered by a pickup failure of a component, is prevented, or a component shortage in a tape is grasped. PATENT LITERATURE Patent Literature 1: JP-A-2014-072409 BRIEF SUMMARY Technical Problem Here, since the brightness of a body portion or the like of a component differs from a component type to a component type, in the device described above, a threshold need be set for each component type, whereby the determination processing becomes complicated and troublesome. In addition, since the brightness of a body portion of a component can be changed depending on imaging conditions even in the same component type, with the determination method in which the brightness of a body portion of a component is compared with a threshold based on the brightness of the body portion or a threshold based on the brightness of a cavity, there is a possibility that an erroneous determination is made by different imaging conditions. A main object of the present disclosure is to determine accurately on existence or non-existence of a component inside a cavity. Solution to Problem The present disclosure employs the following means in order to achieve the main object described above. According to the gist of the present disclosure, there is provided an image processing device for processing an image of a tape having provided therein multiple cavities for receiving a component as a supply target, the image processing device including: a feature amount acquisition section configured to extract brightnesses of pixels within a predetermined range which can include a bottom surface of the cavity and a component inside the cavity using the image and acquire a value indicating scatter of the extracted brightnesses as a feature amount; anda determination section configured to determine on existence or non-existence of a component inside the cavity based on a determination threshold determined based on the feature amount for each of a case in which a component exists inside the cavity and a case in which no component exists inside the cavity and the feature amount acquired from the image. The image processing device of the present disclosure acquires the value indicating the scatter of the brightnesses extracted from the predetermined range of the image as the feature amount and determines on existence or non-existence of a component inside the cavity based on the determination threshold determined based on the feature amount for each of the case in which a component exists inside the cavity and the case in which no component exists inside the cavity and the feature amount acquired from the image. Here, in case a component exists inside the cavity, the scatter of brightnesses becomes large since the bottom surface of the cavity and the component are captured within the predetermined range, whereas in case no component exists inside the cavity, the scatter of brightnesses becomes small since only the bottom surface of the cavity is captured in the predetermined range. As a result, whether component types, imaging conditions, or the like differs, the feature amounts acquired from the image exhibit the same tendency depending on existence or non-existence of a component inside the cavity. Therefore, even though components types, imaging conditions, or the like differs, existence or non-existence of a component inside the cavity can be determined accurately. BRIEF DESCRIPTION OF DRAWINGS FIG. 1 A schematic configuration diagram of mounting device 10. FIG. 2 A schematic configuration diagram of feeder 20. FIG. 3 A block diagram illustrating a configuration regarding a control of mounting device 10. FIG. 4 An explanatory diagram illustrating an example of an identifier 38. FIG. 5 A flowchart showing an example of feature amount learning processing. FIG. 6 An explanatory diagram illustrating an example of a component-existent image. FIG. 7 An explanatory diagram illustrating an example of a component-non-existent image. FIG. 8 An explanatory diagram illustrating an example of relationships between component existence and component non-existence and corresponding feature amounts. FIG. 9 A flowchart showing an example o