リアロップ List of Products and Services

"Flying One" is a RAW camera that is well-suited for drones. With a fisheye lens that has a deep depth of field, no focusing is required. The angle of view exceeds 180 degrees, eliminating the need for composition adjustments. Additionally, it records the space as it is, making it suitable for capturing fast-moving sports. 【Features】 ■ Simple structure - Compact, lightweight, low power consumption, high quality, and low price ■ Fisheye lens - No need for focusing, composition adjustments, or gimbals, resulting in fewer retakes ■ Uncompressed RAW - High quality due to AI super-resolution reduction *For more details, please refer to the PDF document or feel free to contact us.

• Audio and video equipment
• Video Editing
• Video Production

"Optical Learning (R)" is a technology that performs optical simulations based on lens design data and restores images to be free of aberrations by learning from the results. It achieves image quality comparable to professional-grade codecs (based on SN standards). This also leads to reduced burden in lens design and cost savings. 【Features】 ■ Patent No. 6164564 ■ Restores all types of aberrations and distortions caused by MTF in wide-angle lenses, fisheye lenses, endoscopes, etc. ■ Effective for image processing in stitching, CG, VR, image recognition, and more. *For more details, please download the PDF or feel free to contact us.

• others

"AIRD (Artificial Intelligence Retina Development, formerly Phase Shift Development)" is a super-resolution RAW development technology using machine learning. It is well-suited for RAW development for Optical Learning. It eliminates zipper noise and removes blur caused by optical LPF and diffraction. Additionally, it prevents the occurrence of color artifacts, improves the resolution of red, blue, green, and monochrome, and reduces jaggies. 【Features】 ■ Well-suited for RAW development for Optical Learning ■ Removes blur caused by optical LPF and diffraction ■ Reduces sensor noise ■ Suppresses the emphasis on demosaicing degradation *For more details, please refer to the PDF document or feel free to contact us.

• others

"Keep Resolution Mapping (Super Resolution Geometric Transformation)" is a unique technology that performs geometric transformations without reducing resolution using machine learning. "Super Resolution Stepless Zoom" is a stepless electronic zoom utilizing machine learning.

• Video Editing

"Visual Sharpness" is a sharpness feature that changes only the tilt without adding edges, based on a visual model.

• Video Editing

By combining distance measurement methods such as stereo cameras with Optical Learning, we create images that capture both distant and close subjects like the human eye. - To achieve a deeper depth of field, we cannot increase the size of the image sensor optically. - The aperture ratio of the image sensor, even with back-illuminated types, has reached its limit. - The only way to design a camera that has a deep depth of field and high sensitivity while being easy to use is through this approach. - Visual AI realizes an ideal pan-focus with a wealth of information.

• Video Editing
• photograph
• VR/AR

This document discusses the detection of concrete cracks using machine learning and ultra-high-definition images from RAW data. By processing RAW data output from commercially available digital cameras through techniques such as restoration and synthesis using machine learning, we can enhance the detection accuracy of concrete cracks by producing high-definition images that are undistorted, similar to human vision. [Contents (excerpt)] ■ Introduction ■ Image Processing Using Machine Learning ■ Super Resolution Stepless Zoom ■ Keep Resolution Mapping ■ Optical Learning *For more details, please refer to the PDF document or feel free to contact us.

• others