Gli esperti di tutto il mondo si affidano a Mipar per analizzare le loro immagini. Grazie all'integrazione dell'intelligenza artificiale, Mipar è uno degli strumenti più potenti sul mercato. Inoltre, automatizza le tue analisi ed è adatto a tutte le aree che utilizzano immagini.
Faced with tedious and variable image measurements, researchers use MIPAR to detect and measure:
Capable of identifying and measuring characteristics of all types and formats of images:
Life Sciences
| Fluorescence | Histology | Petri dish |
 |
 |
 |
|
Mitochondrial distribution in a differentiated PC12 cell Fully automatic detection of mitochondria, separation of neurites from the cell body, and measurement of the mitochondrial fraction in each. |
H&E staining of odontogenic keratocystic tumor epithelium Quantification of the thickness of the epithelial layer, after fully automated layer identification |
Bacterial colonies on a petri dish |
Materials Sciences
| Size and shape of silica nanoparticles | Porosity and thickness analysis of nanofibres | Analysis of cracks, identification of part defects |
 |
 |
 |
|
Fully automated solution for the characterization of the size and shape of nanoparticles. Quantify a set of samples from a single scan or batch processing. |
Quantify fibre density, void porosity and fibre thickness distribution. |
Quantify the density of additive manufacturing cracks, size distribution, localized density. Fully automated solution for the quantification of a single field or an entire surface point. |
Inspection and control
| Automated inspection of railway tracks using UAV imaging | Aerial study of the severity of deforestation | Contaminant control |
 |
 |
 |
|
Railway track components quantified from UAV images as part of a weekly railway inspection mandated by the Federal Railway Administration (FRA) |
Percentage of wooded (and deforested) areas automatically quantified from UAV imagery |
Classification and size measurement of contaminants in filter paper The detected contaminants are classified into predefined types and the size and shape are reported |
Trace: Trace your characteristics on a few images. Even use the plots you already have
Form: A deep neural network learns to recognize what you've traced.
Run: Run your model on a new image to detect complex features.
Deep Learning fonctions :
Detecting grain by ignoring twins has been a challenge for the community for decades. Mipar's Deep Learning feature has successfully met this challenge.
The complex, overlapping networks of nanofibres hardly contrast with the background.
Screenshot of the batch processor that reveals the layout and purpose of each element of the user interface.
Here's how to set up and run a batch process to segment and measure features from a set of images.
