PDF ke Teks
Ekstrak teks yang dapat dipilih dari PDF di browser Anda. Halaman hasil pindaian memerlukan OCR, yang tidak disertakan.
Baca teks bahasa Inggris dari foto di browser Anda. Penggunaan pertama mengunduh file pengenalan. Foto nyata bisa menghasilkan huruf yang salah.
Alat ini berjalan di browser Anda. Input, file, dan nilai yang dihasilkan tetap di perangkat ini. Favorit dan alat yang baru dipakai, jika Anda menggunakannya, hanya menyimpan nama alat di penyimpanan lokal — tidak pernah kata sandi, dokumen, atau isi kode QR. Lihat kebijakan privasi.
Untuk saat ini, panduan lengkap alat ini (cara pakai, contoh, dan FAQ) masih dalam bahasa Inggris.
This reads English text from one photo in this tab. It is slower than the canvas image tools. The first time, this browser downloads the recognition engine and the English language file, about 6.6 MB. Later runs on this browser reuse that download. The result depends on the photo. A readable sign can still come back with wrong letters, and a busy background can add lines that are not text. Blur, glare, and small type make that worse. Check the text before you use it.
A real outdoor photo of a printed sign, with bark and leaves around it. The hunting and trespassing lines came back. PRIVATE PROPERTY came back missing letters, and VIOLATORS came back as VIOLKTORS. The largest word, POSTED, did not come back as a clean word, and the bark added extra lines.
After you press Read text, this tab loads the recognition engine and the English language file from this site. The photo is drawn smaller when its long side is over 1,600 pixels, then the words are read in this tab. The image is not uploaded.
English only. One image at a time. JPG, PNG, and WebP only, up to 25 MB. The first run downloads about 6.6 MB. Later runs on this browser reuse that download. Reading a photo is slower than the canvas image tools. The result depends on the photo. A readable sign can still come back with wrong letters, and a busy background can add lines that are not text. Blur, glare, and small type make that worse. A long side over 1,600 pixels is reduced before reading.
See the disclaimer for what these tools do not cover.
Ekstrak teks yang dapat dipilih dari PDF di browser Anda. Halaman hasil pindaian memerlukan OCR, yang tidak disertakan.
Kompres gambar JPG, PNG, dan WebP di browser Anda tanpa mengunggahnya.
Ubah ukuran gambar di browser Anda: atur lebar dan tinggi, kunci rasio aspek, lalu unduh.
Seret kotak di atas JPG, PNG, atau WebP dan unduh piksel di dalamnya. Bebas, 1:1, 4:3, atau 16:9. Rotasi dan filter tidak tersedia di sini.
Ubah JPG ke PNG atau PNG ke JPG di browser Anda, dengan warna latar untuk transparansi.
Ubah halaman PDF menjadi gambar JPG di browser Anda. Pilih halaman pertama, halaman tertentu, atau semua halaman.
This reads English text from one photo in this tab. It is slower than the canvas image tools. The first time, this browser downloads the recognition engine and the English language file, about 6.6 MB. Later runs on this browser reuse that download. The result depends on the photo. A readable sign can still come back with wrong letters, and a busy background can add lines that are not text. Blur, glare, and small type make that worse. Check the text before you use it.
Use one JPG, PNG, or WebP image. A long side over 1,600 pixels is reduced before reading. English is the only language.
Drag and drop a photo here, or choose a file.
One image, up to 25 MB.
The image stays in this tab. It is not uploaded. The recognition files are loaded from this site.