Magnetic material appearance size sorter
The Boming magnetic material appearance size sorting equipment uses deep learning to enhance feature extraction, maintaining a strong defect detection capability even when defect features are not obvious. It also uses traditional image processing to extract defect characteristics, allowing it to more closely meet customer-defined judgment conditions, providing quantitative judgment and enabling quick modification of judgment conditions. The operation is simple and clear; with a single-button start, it performs comprehensive detection of defects such as size, chipped corners, hidden cracks, and poor plating on the magnetic materials.
Core Advantages
Intelligence
Integrates multiple technologies, powerful hardware and software
(Complete range of standard and customized products)
Advanced Algorithms
Deep Learning defect detection algorithm, model inference (defect segmentation or defect location)
(Background interference removal, defect merging, detected defect screening, etc.)
One-Click Startup
One-click detection startup, multiple cameras work independently
(Numbering summary sends detection results)
Automation
Automatic detection, products flow out with the conveyor line, no manual selection required
(Detection workstation optional)
High Precision
High dimensional measurement accuracy, fast speed, high detection efficiency
(Repeat detection accuracy: ±5μm)
Customizable Development
Customizable development according to actual needs,
(Customized Development)
Vision Solution
High intelligence
Software Solution
· Based on the existing turntable machine general software, customized modifications are made specifically for this magnet.
Different camera modules are assigned according to different detection vision or detection items.
· One-click detection startup, multiple cameras work independently, and the number summary sends detection results.
· Compatible with traditional version and deep version detection algorithms.
· Provides a defect type detection whitelist function to selectively detect different defects.
· Provides multi-level access control to ensure the encryption of software parameter modifications.
Algorithm Solution
Dimensional detection uses traditional measurement algorithms
Traditional image processing method preprocessing
Target validity determination
Background interference removal
Feature enhancement
Dimensional transformation
· Defect detection algorithm uses traditional image processing algorithm and deep learning defect detection algorithm
Deep Learning defect detection algorithm
Model inference (defect segmentation or defect location)
Self-developed customized algorithm
The detection workstation is optional and can be configured according to actual needs. The equipment size will vary with different configurations.
Defect Detection Effect
The detection workstation is optional and can be configured according to actual needs. The equipment size will vary with different configurations.
Solution Parameters
Equipment size (L*W*H)
2660*1400*1800mm
Detection product size
Length: 15-55mm Width: 4-30mm Thickness: 0.5-5mm
Detection product shape
Squares, tiles
Detection product plating
Nickel plating, black epoxy
Dimensional detection items
Length, width, thickness, parallelism, perpendicularity, R angle
Dimensional detection accuracy
Repeat detection accuracy: ±5μm
Appearance defect detection items
Mechanical defects: missing corners, hidden cracks, warping, internal cracks, knife marks
Plating defects: particles, shrinkage, delamination, watermark, fluff
Minimum detectable size of appearance defects
Length * width: 0.06 * 0.06mm
Magnetic direction
Detects the magnetic direction, the miss rate is 0, optional
Detection efficiency
100-200pcs/min
Video Number
Public Account
Douyin Signal
Zhuhai Bomming Vision Technology Co., Ltd SEO Privacy Policy
Business License