This page provides a list of frequently asked questions by customers
Currently, the platform supports common image file formats, including:
JPEG (.jpg, .jpeg) 🖼️
PNG (.png) 🖼️
BMP (.bmp) 🖼️
GIF (.gif) 🖼️
TIFF (.tif, .tiff) 🖼️
For batch analysis, you can compress multiple image files into a single ZIP archive 📂 and upload it to the platform.
💡 Friendly reminder: If your files are in other formats (such as DICOM, commonly used for medical imaging), please convert them to one of the supported image formats listed above before uploading.
Token calculation methods vary depending on the analysis type:
A. Model Inference (Prob) and Image Segmentation (Lbls):
Calculated based on the total number of pixels in the image:
Token number = total number of pixels ÷ (2¹⁰ × 2¹⁰)
That is, every 1,048,576 pixels are counted as 1 token.
Example 📊: A standard 1024 × 1024 pixel image typically consumes 1 token.
B. Geometric Statistics (Stat) Analysis (e.g., object size, shape, etc.):
Calculated based on the number of objects after segmentation:
**Token number = total number of objects ÷ (2¹⁰ × 2³) **
That is, every 8,192 objects are counted as 1 token.
The platform offers a free trial 🎁, available once every 14 days per email account, and valid for 30 days upon activation.
Each trial includes 8 tokens, which can be used for various types of image analysis.
Trial Example 💡:
Using the "Model Inference (Prob)" function, a standard 1024 × 1024 pixel image typically consumes 1 token, meaning your trial can process approximately 8 standard-sized images.
⚠️ Please note: The actual number of images you can analyze depends on both the image resolution and the type of analysis selected. For example, geometric statistics (Stat) use a different token calculation method based on the number of segmented objects.
As long as your account's token balance remains positive (greater than 0), you can continue using the platform for image analysis.
✨ Trial Period Grace Behavior:
During the trial period, even if a specific analysis requires more tokens than your remaining balance, the platform will still complete the analysis and provide the result file for download.
⚠️ Please note: This grace behavior applies only during the trial period. The platform reserves the right to monitor and restrict usage in cases of abuse.
If you anticipate needing more tokens for continued use, we recommend upgrading your quota 📈 or contacting us for a tailored solution.
🔁 Batch Analysis (2D Models Only)
Currently, the platform’s batch analysis feature supports only 2D image models.
To analyze multiple images at once, please follow the steps below:
1. Prepare Your Images 🗂️
Place all the image files you wish to analyze into a single folder.
✅ Supported formats: JPG, PNG, TIFF, and other common image types.
2. Compress the Folder 🗜️
Zip the entire folder to create a single .zip file containing all the images.
3. Upload to the Platform 📤
Go to the upload section of the platform and submit your .zip file.
The system will automatically process each image inside and perform batch analysis.
4. Download the Results 📦
Once analysis is complete, all output files will be packaged into a downloadable ZIP file, including:
📷 Processed image results (TIFF)
📊 CSV statistics (if applicable)
📄 HTML report summarizing the analysis
The platform does not impose strict limits on the size of image files you upload 📏.
However, if a single file exceeds the real-time processing capacity of the web interface, the system will automatically upload it to Google Cloud temporary storage ☁️ to enable smooth access and processing by the backend computing servers.
To protect your privacy and data security 🔒:
All uploaded files are automatically deleted within 24 hours.
If you prefer to delete your files immediately after upload, please contact our customer support team 📞—we’ll be happy to assist you.
The output you receive will depend on the analysis type you selected. Details are as follows:
1. Model Inference (Prob):
You will receive a TIFF image file 📷 generated by the model, typically containing a probability map or raw model output.
2. Image Segmentation (Lbls):
In addition to the inference result (TIFF), you'll receive another TIFF file 🖼️ representing the segmentation result, where each region is assigned a unique label ID to support downstream analysis or visualization.
3. Geometric Statistics (Stat):
On top of the inference and segmentation images, you will also receive a CSV file 📊 containing geometric measurements and feature values for each segmented object, such as area, perimeter, shape factor, and more.
All analysis results are packaged into a downloadable ZIP file 📦, which includes:
📷 TIFF image files (inference and/or segmentation)
📊 CSV file (only for Stat analysis)
📄 HTML report: summarizes the analysis process, computation time, and result overview for convenient reference and archiving.
To use the platform's 3D analysis model ⚛️, please prepare and upload your image data in one of the following recommended formats:
✅ Recommended Formats:
A single 3D TIFF file 📁, containing your volumetric data.
Or, a set of sequential 2D slice images (e.g., PNG, TIFF) 💾 arranged in proper order, then compressed into a ZIP file for upload.
Make sure the 2D slices are named in a way that preserves their correct stacking order (e.g., slice_001.tif, slice_002.tif, ...).
⚠️ Important Note:
Due to current system resource configurations, 3D models do not support batch analysis like 2D models 🚫. Each analysis is limited to one 3D dataset per upload. Please plan your workflow accordingly.
You can apply for free tokens through our Free Trial Page by following these steps:
📧 1. Submit Your Email
Enter your email address on the trial request form.
📩 2. Check Your Inbox
You will receive an activation email from:
webadm-noreply@chingyeh.com
⚠️ Please add this address to your safe sender list to avoid the email being marked as spam.
🔗 3. Activate Your Trial
Click the "Activate Now" link in the email to confirm your registration.
Once activated, your account will be granted the free trial quota, and you can begin using the platform immediately.