<div dir="ltr">Just to confirm what Steven said, building from current master and applying your command to your `nuclei.jpg` image I obtain the attached image without artifacts.<div><br></div><div>0.11 should be released fairly soon, but if you need this sooner you can build from source. We'd be happy to help if you run into problems.</div><div><br></div><div>Josh</div><div><br><br>On Saturday, January 24, 2015 at 4:46:23 AM UTC-7, Kai Wiechen wrote:<blockquote class="gmail_quote" style="margin: 0;margin-left: 0.8ex;border-left: 1px #ccc solid;padding-left: 1ex;"><div dir="ltr"><p style="text-align:center;clear:both"><a style="clear:left;margin-bottom:1em;float:left;margin-right:1em" href="https://lh3.googleusercontent.com/-JDkNXe7k44s/VMOE-xwgPXI/AAAAAAAABLA/STWxQlQkU9Q/s1600/nuclei.jpg" target="_blank" rel="nofollow" onmousedown="this.href='https://lh3.googleusercontent.com/-JDkNXe7k44s/VMOE-xwgPXI/AAAAAAAABLA/STWxQlQkU9Q/s1600/nuclei.jpg';return true;" onclick="this.href='https://lh3.googleusercontent.com/-JDkNXe7k44s/VMOE-xwgPXI/AAAAAAAABLA/STWxQlQkU9Q/s1600/nuclei.jpg';return true;"><img border="0" origsrc="https://lh3.googleusercontent.com/-JDkNXe7k44s/VMOE-xwgPXI/AAAAAAAABLA/STWxQlQkU9Q/s1600/nuclei.jpg"></a><a style="margin-left:1em;margin-right:1em" href="https://lh3.googleusercontent.com/-sgqpXIXHhF0/VMOFDbcK8sI/AAAAAAAABLI/iHcFERpbFfc/s1600/nuclei_clahe.jpg" target="_blank" rel="nofollow" onmousedown="this.href='https://lh3.googleusercontent.com/-sgqpXIXHhF0/VMOFDbcK8sI/AAAAAAAABLI/iHcFERpbFfc/s1600/nuclei_clahe.jpg';return true;" onclick="this.href='https://lh3.googleusercontent.com/-sgqpXIXHhF0/VMOFDbcK8sI/AAAAAAAABLI/iHcFERpbFfc/s1600/nuclei_clahe.jpg';return true;"><img border="0" origsrc="https://lh3.googleusercontent.com/-sgqpXIXHhF0/VMOFDbcK8sI/AAAAAAAABLI/iHcFERpbFfc/s1600/nuclei_clahe.jpg"></a></p>Hello,<br><br>I am new to scikit-image (and python too), but found the combination of scikit-image implementation of algorithms together with ndarrays extremely powerful.<br><br>For now, I am trying to find nuclei in microscopic images (color deconvolution, find seeds with blob_log) This works fairly well but due to low contrast some nuclei are not detected. <br><br>nuclei_clahe = equalize_adapthist(nuclei) <br><br>gives much better local contrast but introduces linear artefacts at the test image borders. Is it possible to avoid/reduce this?<br><br>Best regards,<br><br><br>Kai<br><br><br><br><br><br></div></blockquote></div></div>