Seeing the Defects Cameras Miss: Photometric Stereo Inspection at the Kariya Automotive ExhibitionNew!!
TOMOMI RESEARCH exhibits at the Kariya automotive exhibition, Nov 25–26, 2026 (Booth No. 19), showing how photometric stereo reveals surface defects standard cameras miss for AI inspection.
Anomaly Detection Trained on Good Parts — Visual Inspection Fundamentals: 10 Key QuestionsNew!!
Anomaly detection for visual inspection in 10 questions: training on good parts, anomaly score thresholds, miss and false positive rates, line operation.
Photometric Stereo on Artificial Leather: Color vs. Grain ShapeNew!!
We imaged embossed red artificial leather with TR-100 (FORESIGHT STEREO®). One capture separates the color pattern from the grain's shape, revealing thin grooves where color shows broad bands.
Same Face, Two Views: Bringing Inspection-Grade Imaging to a Community Wellbeing Festival at Shonan iParkNew!!
On November 21, 2026, TOMOMI RESEARCH, INC. brings FORESIGHT STEREO®, a lighting technology built for industrial surface inspection, to the Shin-Shonan Wellbeing Festa 2026 at Shonan iPark. Visitors can see their own face in two views: reflectance and fine surface relief.
Launch of E3 Anomaly Synthetic Generator
On October 27, 2026, TOMOMI RESEARCH Inc. released E3 Anomaly Synthetic Generator, software for creating synthetic NG data. From good (OK) images and texture images, it creates synthetic NG images containing defects synthesized by this software, paired with pixel-level ground-truth masks.
We Exhibited at a Technology Exhibition and Business Meeting at Suzuki Headquarters
On October 14, 2026, we exhibited at the FY2026 Kanagawa Automotive-Related Technology Exhibition and Business Meeting in SUZUKI, held in the Collaboration Space on the 1st floor of the West Building of Suzuki Motor Corporation in Hamamatsu, Shizuoka Prefecture. The event was organized by Kanagawa Prefecture with the support of the Kanagawa Industrial Promotion Center (KIP).
Logical Anomaly Detection Rankings Flip When Artificial Capture Shifts Are Mixed In
In visual inspection on production lines, AI finds defects by looking for differences from good parts. A new paper shows that methods scoring near-perfect AUROC on MVTec LOCO change places once artificial changes in brightness and image scale are mixed into the test.
Creating Pseudo-Defects "Just Outside Normal"
Visual inspection AI in factories is often trained on good parts only, so plausible defects are created artificially from normal images. GLASS (ECCV 2024) reports improved detection of faint and small defects by refining how these pseudo-defects are created.
Seeing a Populated Circuit Board as "Color" and "Shape": PCB Visual Inspection and Four Image Representations
In visual inspection of populated circuit boards, specular reflections from solder and leads often make "shape" hard to read. We captured the control board of a commercially available HDD with TR-100 (FORESIGHT STEREO®) and compared the four image representations obtained from a single capture.
Featured in the October Issue of the Monthly Magazine "Columbus" | Why Scratches on Glossy Surfaces Are Hard to See, and Why AI Visual Inspection Does Not End at Installation
Featured on p.42 of the October 2026 issue of the monthly magazine "Columbus." We explain why scratches on glossy surfaces are hard to see, the principle behind the 3D visual inspection system TR-300, and why introducing AI visual inspection requires several months of tuning.