A method includes: determining a defective area in a semiconductor device of a semiconductor wafer; thinning the semiconductor wafer from a backside of the semiconductor wafer; bonding a first substrate to the backside of the semiconductor wafer, wherein the first substrate includes an opening and the defective area is exposed through the opening; and performing a test on the defective area by projecting a light beam from the backside through the opening.

The purpose of this invention is to make it possible to efficiently and accurately detect a reticle defect at an earlier stage in a manufacturing process. The inspection device according to this invention uses the results of comparing die images of wafers that have had patterns transferred thereto using the same reticle after subjecting the die images to averaging processing and the results of comparing the die images without subjecting the same to averaging processing to distinguish a random defect signal caused by a huge defect, or the like, and having an extremely high brightness from a repeated defect signal, and only extracts repeated defects with higher accuracy.

A component mounting device includes a mounting head that mounts a component at a mounting position on a substrate, and a control unit that determines a mounting state of another component mounted in advance around the mounting position based on height information of a region around the mounting position.

An image acquiring method, an image acquiring apparatus and a wafer inspection apparatus are disclosed. A line scan camera is disposed above a transfer path of a wafer to continuously acquire partial images having a predetermined size by imaging a scan area including a portion of the transfer path, and the partial images are stored in an image storage unit. A partial image including a predetermined feature point among the partial images is detected by an image analysis unit, and an image merging unit merges a predetermined number of partial images including the detected partial image to acquire an entire image of the wafer. An image inspection unit analyzes the entire image of the wafer to detect defects in the wafer.

An inspection system of semiconductor device includes a light source for producing a light beam, a lens module including at least one metalens, a receiver, and a processor. During an inspection process, the light beam emitted from the light source is focused on a target object by the metalens of the lens module and is reflected to form a reflected light by the target object. The receiver is used for receiving the reflected light. The processor is used for receiving an electric signal corresponding to the reflected light and generating an inspection result.

A printed circuit board inspection apparatus can inspect the mounting state of a component by generating depth information on the component mounted on a printed circuit board by using a pattern of light reflected from the component and received by an image sensor, inputting the generated depth information into a machine-learning-based model, obtaining depth information with reduced noise from the machine-learning-based model, and using depth information with reduced noise.

An apparatus, a method and a computer program product for defect detection in work pieces is disclosed. At least one light source is provided and the light source generates an illumination light of a wavelength range at which the work piece is transparent. A camera images the light from at least one face of the work piece on a detector of the camera by means of a lens. A stage is used for moving the work piece and for imaging the at least one face of the semiconductor device completely with the camera. The computer program product is disposed on a non-transitory, computer readable medium for defect detection in work pieces. A computer is used to execute the various process steps and to control the various means of the apparatus.

An optical inspection is performed to detect potential defects within integrated circuit devices and a first electron-based inspection of less than all of the potential defects is performed to identify primary actual defects. A process window of manufacturing parameter settings used to manufacture the integrated circuit devices is identified and the integrated circuit devices manufactured using the manufacturing parameter settings inside the process window have less than a threshold number of the primary actual defects. To identify additional actual defects a second electron-based inspection is performed that is limited to selected ones of the potential defects in the integrated circuit devices that were manufactured using the manufacturing parameter settings inside the process window but were uninspected in the first electron-based inspection.

Various defects in an electronic assembly can be intelligently identified with a system having at least a server connected to a first capture module and a second capture module. The first capture module may be positioned proximal a first manufacturing line while the second capture module is positioned proximal a second manufacturing line. Images can be collected of first and second electronic assemblies by respective first and second capture modules prior to the images being sent to a classification module of the server where at least one defect is automatically detected in each of the first and second electronic assemblies concurrently with the classification module.

An optical inspection apparatus includes an interferometer module, which is configured to direct a beam of coherent light toward an area under inspection and to produce a first image of interference fringes of the area. The apparatus also includes a triangulation module configured to project a pattern of structured light onto the area, and at least one image sensor configured to capture the first image of interference fringes and a second image of the pattern that is reflected from the area. Beam combiner optics are configured to direct the beam of coherent light and the projected pattern to impinge on the same location on the area. A processor is configured to process the first and second images in order to generate a 3D map of the area.