Spot Product Lineup

【Electronic Component Mounting】Determination of solder defects on the substrate. Detects bridging, insufficient solder, and lifting. 【Metal Pressing】Surface scratches and dents. Detects points that deviate from the normal surface standard. 【Resin Molding】Burrs and chips. Detects abnormalities in the gate area and parting surface. 【Film】Foreign objects. Detects against the clear standard of good products. 【Assembly】Missing parts and assembly errors. Detects the presence and position of components. 【Printing】Stamping defects. Detects misalignment and omissions of characters. For all targets, we first conduct a free model creation (first time free) to verify accuracy using actual images, and only after meeting the standards do we proceed with implementation. By incorporating a DLL into existing equipment, processes with few defective product samples can also be addressed with unsupervised anomaly detection.

• Image Processing Software
• Visual Inspection Equipment

In processes where defects rarely occur, it is not possible to collect defective product samples. The DeepAge Sensor's unsupervised anomaly detection learns the standard of "normal" using only good product images and detects images that deviate from that standard. It does not ask for "hundreds of defective product images." You can start inspection operations from the early stages of implementation, and it can also identify defects that have never been seen before as "deviating from normal." Inference operates on the CPU (no GPU required), and integration into the device is done with just one DLL. Later, if defective product images accumulate on-site, you can transition to a supervised model with the same API structure. In processes where samples are insufficient, you can first verify accuracy with free model creation.

• Image Processing Software
• Visual Inspection Equipment

Are you relying on visual inspections, leading to subjective judgments? The "experienced eye" is lost due to retirements and transfers, and securing inspectors is not easy. DeepAge Sensor is an AI image inspection software that automates visual inspections. It starts learning with only good product images, making it feasible to implement even in processes with only a few defective samples. It can be integrated into existing inspection equipment by simply adding one inspection DLL and runs on industrial PCs without the need for a GPU. You can first try creating a model for free using your inspection images and decide on implementation after reviewing the accuracy verification results, so the risk of consideration is almost nonexistent.

• Image Processing Software
• Visual Inspection Equipment

To device manufacturers and system integrators who want to incorporate AI inspection functions into devices that handle images: DeepAge Sensor is provided as a pre-built inference library (Windows/UNIX). You only need to reference a single public C header (extern "C"). It works simply by linking in C/C++ or loading with ctypes in Python. The CPU version requires no GPU, CUDA, or drivers, and operates at 12ms per image (the GPU version operates at 2ms). Training is conducted on our cloud, and only the model file is distributed to the devices, so there is no need to rebuild the device software every time the model is updated. An unsupervised anomaly detection library (.uai) can also be used with the same API structure. We will provide a library and sample model for evaluation, so you can first check if create → load → inference works in your environment. OEM/white label offerings are available. Your company will handle first-level support, while we will handle second-level support.

• Image Processing Software
• Visual Inspection Equipment
• Software (middle, driver, security, etc.)

We are engaged in development centered around deep learning, which has recently attracted significant attention among artificial intelligence technologies. Of course, the technologies we can provide in response to customer requests vary widely. We accurately capture needs and provide optimal solutions. What we can do: Image recognition, natural language processing, speech recognition, recommendation engines, anomaly detection using sensor data, customer clustering, optimization of direct mail and advertising, matching optimization, market forecasting, optimization of product ordering and inventory, fault detection, LTV (customer lifetime value) prediction, real estate forecasting, site conversion analysis, survey data analysis, inquiry analysis, consulting and requirements definition support related to artificial intelligence technologies, and other systems related to artificial intelligence and machine learning in general.

• Other information systems
• Other operation management software
• Other network tools

The DeepAge Sensor is a high-precision defective product detection sensor that uses artificial intelligence. It combines three elements: high precision, high speed, and easy implementation. It can be added to existing systems regardless of whether they are Windows or Unix, so there is almost no additional implementation cost required. If you have concerns about the accuracy or speed of your existing system, or if you are considering implementing a new system, you can easily incorporate artificial intelligence. 【Detailed Materials and Inquiries】 Please feel free to contact us if you would like to request detailed materials. If you have requests for more detailed information for future reference, please do not hesitate to reach out.

• Image Processing Software
• Visual Inspection Equipment